#gemrb@irc.freenode.net logs for 22 Jan 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:06:16] <brada> I would think i could reproduce a crash at least once in a while
[00:07:13] <-- WingedHussar has left IRC (Ping timeout: 245 seconds)
[00:10:24] <-- edheldil has left IRC (Ping timeout: 240 seconds)
[00:12:43] <brada> oh well. pick this up again on the marrow
[00:12:47] <-- brada has left IRC (Quit: brada)
[00:47:10] --> WingedHussar has joined #gemrb
[00:54:22] --> brada has joined #gemrb
[01:46:39] --> Canageek has joined #gemrb
[02:26:42] --- ermo is now known as ermo^
[04:04:40] <-- brada has left IRC (Quit: brada)
[04:21:19] --> brada has joined #gemrb
[06:01:01] <-- brada has left IRC (Quit: brada)
[06:49:58] --> edheldil has joined #gemrb
[06:49:58] --- ChanServ gives channel operator status to edheldil
[06:54:46] <-- edheldil has left IRC (Ping timeout: 256 seconds)
[07:11:49] --> miha has joined #gemrb
[07:43:02] --> fuzzie_ has joined #gemrb
[07:43:12] <-- fuzzie has left IRC (Ping timeout: 276 seconds)
[07:48:04] <-- Canageek has left IRC (Ping timeout: 248 seconds)
[08:02:07] --- fuzzie_ is now known as fuzzie
[09:01:23] --> lynxlynxlynx has joined #gemrb
[09:01:23] <-- lynxlynxlynx has left IRC (Changing host)
[09:01:23] --> lynxlynxlynx has joined #gemrb
[09:01:23] --- ChanServ gives channel operator status to lynxlynxlynx
[10:01:03] --> hadrian has joined #gemrb
[10:16:00] --> edheldil has joined #gemrb
[10:16:01] --- ChanServ gives channel operator status to edheldil
[10:25:33] <hadrian> Hi, I just came across GemBR and I love the idea, can’t wait for 1.0
[10:39:53] <lynxlynxlynx> :)
[10:39:59] <lynxlynxlynx> what will happen at 1.0?
[10:44:55] <hadrian> It will have implemented its awesome ideas, I suppose.
[10:48:44] <hadrian> (Which does not mean that I won’t try it out earlier)
[10:55:22] <lynxlynxlynx> cool, they already tried gemrb on raspberry
[11:00:40] <Seniorita> [wiki] engine:platforms http://www.gemrb.org/wiki/doku.php?id=engine:platforms&rev=1358852138&do=diff
[11:26:02] <-- Drakkar has left IRC (Ping timeout: 255 seconds)
[11:33:32] --> Drakkar has joined #gemrb
[12:05:07] <traveler__> hah at lynx: 'what will happen at 1.0?' :)
[12:05:32] <miha> "what will happen just before 1.0" ?
[12:06:15] <lynxlynxlynx> miha will make a new apk
[12:06:28] <lynxlynxlynx> check the last message here: https://sourceforge.net/tracker/?func=detail&atid=360122&aid=3599172&group_id=10122
[12:06:30] <Seniorita> SourceForge.net: GemRB Game Engine: Detail: 3599172 - Android Build
[12:06:35] <traveler__> are we at 0.9.9 then :P ?
[12:06:36] <lynxlynxlynx> and cry a little :D
[12:07:12] <lynxlynxlynx> traveler__: actually i never managed to get everyone to agree on a roadmap
[12:07:28] <lynxlynxlynx> we do agree that at 1.0 all games should be completable
[12:08:08] <traveler__> re rasp: cool, though they think that editing *cfg is the hard part...
[12:08:40] <traveler__> number ver doesn't matter to me at all
[12:09:01] <traveler__> but as you can see, there are still people attaching magical properties to it
[12:10:11] <miha> lynxlynxlynx: ok, will try :)
[12:10:25] <lynxlynxlynx> thanks :)
[12:11:12] <miha> lynxlynxlynx: will need help with config/demo game, i guess, never played it :)
[12:11:16] <lynxlynxlynx> traveler__: also many instructions for setting up gemrb mention to change GameName ;)
[12:11:31] <traveler__> glancing and thinking 'cool will try when they are not alpha' not realizing all work that already went into it. gemrb has few years, yet is sometimes downplayed just because of ver number
[12:11:44] <lynxlynxlynx> miha: you can run the bunled minimal test, should be good enough
[12:13:12] <hadrian> Well no. I did not presume this is alpha.
[12:13:39] <traveler__> it can get to downright funny lengths though... as mplayer saga with <1 versions for years. but about that all knew, at least.
[12:14:09] <hadrian> I play on Windows so there is no special reason to switch to gemBR other than its extended modding capabilities
[12:14:30] <traveler__> well
[12:14:35] <miha> lynxlynxlynx: i'm git noob, so how to "checkout it to ac1f2e9 revision"
[12:14:37] <traveler__> that's a reason right there ;)
[12:14:44] <lynxlynxlynx> traveler__: think of e17 :)
[12:14:52] <miha> lynxlynxlynx: i know git clone :p
[12:15:06] <lynxlynxlynx> miha: git checkout ac1f2e9 # after you cloned it
[12:15:36] <traveler__> hadrian: you could check for differences/quirks also, pretty nice to catch them running side by side
[12:15:59] <miha> lynxlynxlynx: change project/jni/application/src symlink to ballfield application (ballfield what? gemrb?)
[12:16:09] <traveler__> well, if i would still use windows i would try gemrb just out of curiosity, but that's me
[12:16:17] <hadrian> and while there are some really nice mod ideas (more multicasses, >6 party members) those are not yet implemented
[12:16:36] <hadrian> Yes, I plan to do that :)
[12:16:41] <lynxlynxlynx> monk/sorcerer is :)
[12:16:46] <lynxlynxlynx> good for quick runs
[12:17:09] <lynxlynxlynx> actually i'm looking for iwd1 testers, specifically for the two extensions
[12:17:10] <traveler__> oh
[12:17:18] <lynxlynxlynx> miha: any subdir named like that?
[12:17:24] <hadrian> Yes, but from what I’ve gathered it requires manual work for windows builds
[12:17:24] <traveler__> and gemrb don't have artificial xp cap
[12:18:06] <hadrian> I suppose gemRB can be used with modded versions of BG2?
[12:18:14] <lynxlynxlynx> sure
[12:19:14] <traveler__> re: hadrian you have buildbot binaries for windows available
[12:19:56] <lynxlynxlynx> the manual work is just editing some text files
[12:20:01] <traveler__> and releases as well http://sourceforge.net/projects/gemrb/files/GemRB%20Win32%20Binaries/
[12:20:03] <Seniorita> GemRB Game Engine - Browse /GemRB Win32 Binaries at SourceForge.net
[12:20:24] <lynxlynxlynx> now more of it can be converted to weidu
[12:20:28] <hadrian> I will take a look at this. Maybe I’ll succeed in changing the monk/sorcerer mod to Kensai/sorcerer :)
[12:20:50] <hadrian> Its the class descriptions if I am right …
[12:21:39] <lynxlynxlynx> hmm
[12:22:01] <lynxlynxlynx> make a fighter/sorcerer multi
[12:22:19] <lynxlynxlynx> a few more tables need to be edited if you want to dualclass
[12:22:26] <lynxlynxlynx> otherwise you'd have to hack the kit in
[12:24:38] <hadrian> I might have time to do try this at the weekend
[12:24:41] <hadrian> ;)
[12:25:18] <hadrian> (well some weekend, maybe not this weekend)
[12:39:56] --> kida has joined #gemrb
[12:44:58] <edheldil> there's a tantalizing note "Icewind dale works quite well in 640*480 fullscreen mode but i am experiencing some troubles with the parsing of text dialog" at the raspberry site
[12:47:13] <lynxlynxlynx> maybe he has sight problems
[12:47:34] <lynxlynxlynx> i doubt it was meant in a code sense
[12:48:50] <edheldil> ahhh, that makes sense
[12:53:36] <lynxlynxlynx> hadrian: i eliminated a whole step for the monk/sorcerer mod
[12:53:57] <lynxlynxlynx> checking now if i can weidufy the whole thing
[12:54:17] <lynxlynxlynx> it has variables, so it should be possible *knock knock*
[13:16:22] <hadrian> @lynxlynxlynx I wouldn’t have expected you to put any effort in my fighter/sorcerer ambition, but it sure is highly appreciated!
[13:17:52] <lynxlynxlynx> the cleanup was on my todo anyway
[13:18:06] <lynxlynxlynx> you'll still have to do plenty of data digging anyway
[13:26:23] <traveler__> lynx: the original superfluous window manager? ;)
[13:41:08] <-- edheldil has left IRC (Ping timeout: 248 seconds)
[13:46:47] --> edheldil has joined #gemrb
[13:46:47] --- ChanServ gives channel operator status to edheldil
[14:06:49] <lynxlynxlynx> hadrian: i think i got it
[14:07:03] <lynxlynxlynx> i only need a bg2 tester to verify, it's pure weidu now
[14:12:18] <lynxlynxlynx> https://github.com/lynxlynxlynx/gemrb-mods/tree/master/sorcerer-monk
[14:12:19] <Seniorita> gemrb-mods/sorcerer-monk at master · lynxlynxlynx/gemrb-mods · GitHub
[14:13:16] <lynxlynxlynx> need to start symlinking static game data to save hdd space
[14:15:06] <edheldil> I do that to create the foreign language versions
[14:24:39] --- ermo^ is now known as ermo
[14:31:07] <hadrian> @lynxlynxlynx: Will do. It seems to me, starting from that, a fighter/sorcerer would not be too difficult to create. Thank you
[14:31:39] <lynxlynxlynx> just the table data needs to be changed
[14:32:20] <hadrian> :)
[14:47:13] --> brada has joined #gemrb
[14:50:03] <brada> didnt have any luck compiling for android
[14:50:18] <brada> tons of libstdc++ errors
[14:58:40] <fuzzie> with the instructions?
[15:01:02] <lynxlynxlynx> with clang?
[15:02:09] <-- hadrian has left IRC (Quit: Page closed)
[15:04:08] <fuzzie> pretty sure the crystax ndk isn't supported any more..
[15:06:32] <brada> i think its maybe something to do with c vs c++
[15:06:46] <brada> its looking for things like ::malloc instead of malloc
[15:09:03] <brada> any luck with the fonts?
[15:09:57] <fuzzie> a failure at link time?
[15:11:50] <brada> no it says ::malloc was not declared
[15:12:03] <brada> repeat infinate times replacing malloc with something else
[15:12:09] <fuzzie> which ndk?
[15:12:17] <lynxlynxlynx> i gdbed dbstrlen
[15:12:26] <lynxlynxlynx> seemed to work for normal strings
[15:12:42] <lynxlynxlynx> but it got fed a lot of garbage some times
[15:12:51] <brada> from?
[15:12:55] <lynxlynxlynx> 1-3 char garbage
[15:12:59] <lynxlynxlynx> didn't check
[15:13:19] <lynxlynxlynx> i can revisit that, but now i need to prepare some intake
[15:13:26] <-- miha has left IRC (Quit: Lost terminal)
[15:13:40] <brada> fuzzie: just using latest. really that diffrent from other versions?
[15:14:10] <fuzzie> the ndk changes a lot
[15:14:35] <fuzzie> and the STL libraries have always been somewhat experimental
[15:15:03] <brada> :/
[15:15:34] <fuzzie> but latest ndk (8d) + pelya's stuff should thereotically be ok
[15:16:40] <fuzzie> did you manage to do something about the states btw?
[15:17:30] <fuzzie> i realised it's more complicated than I sleepily thought last night
[15:20:24] <brada> how so?
[15:20:28] <-- traveler__ has left IRC (Ping timeout: 245 seconds)
[15:20:41] <brada> seems to works as far as ive seen just by setting multibyte to false for them
[15:20:52] <fuzzie> because the relevant chars are embedded directly in strings rendered in a different font, i thought
[15:20:54] <lynxlynxlynx> GemRB::Font::CalcStringWidth called from printfromline like valgrind already confirmed
[15:21:24] <fuzzie> just handled with [initials=2], so it seems you'd already have mangled the state glyph by the time you got there
[15:23:29] <brada> lynx: which one?
[15:23:33] <brada> there are 2
[15:23:43] <brada> one takes char the other ieword
[15:23:46] <fuzzie> your valgrind log was complaining before that, no?
[15:24:01] <lynxlynxlynx> Font.cpp:465
[15:24:19] <lynxlynxlynx> though where i'm now, all the strings are within 255
[15:25:07] <brada> i dont see any problem
[15:26:35] <lynxlynxlynx> i remember what i was looking at, it was GetDoubleByteString
[15:27:39] <lynxlynxlynx> http://paste.debian.net/227449/
[15:27:40] <Seniorita> debian Pastezone
[15:28:15] <brada> that doesnt mean anything to me
[15:28:25] <fuzzie> there's nothing else relevant in the backtrace?
[15:28:45] <fuzzie> because that seems to just be a corrupt string
[15:28:52] <brada> yeah
[15:29:58] <brada> fuzzie: do you have this problem?
[15:32:42] <fuzzie> lynxlynxlynx: hmm, do you have message window logging on?
[15:33:23] <fuzzie> brada: can't try it atm.
[15:34:25] <brada> i cant see how my changes can corrupt the string. valgrind didnt report any invalid writes
[15:35:24] <lynxlynxlynx> fuzzie: no
[15:35:57] <lynxlynxlynx> brada: invalid reads can result in that as easily
[15:36:17] <fuzzie> why the invalid reads though?
[15:36:37] <brada> well because its not a real sting
[15:36:48] <brada> so it just keeps going until 0
[15:36:51] <lynxlynxlynx> x> http://sprunge.us/gAZV
[15:37:22] <fuzzie> well, yes, that backtrace snippet makes it seem like it's not the font code
[15:39:18] <brada> no but there is something about depending on an uninitialized value
[15:39:25] <fuzzie> your TextArea changes maybe
[15:39:43] <brada> the one piece of code i kept from the original patch :p
[15:40:08] <fuzzie> also don't see anthing likely there though
[15:41:20] <brada> well they are coming from non font related code
[15:41:50] <fuzzie> well, I wonder what the string is like before/after the SetupString call around TextArea.cpp:729
[15:42:17] <brada> i assume its already garbage
[15:42:24] <brada> but yes good question
[15:43:10] <brada> I need to acquire a non multibyte non english language pack
[15:43:52] <fuzzie> well to me it looks suspiciously like SetupString has the potential to break it
[15:44:00] <brada> lynx: do you have english versions?
[15:45:19] <lynxlynxlynx> only english
[15:45:35] <brada> hmmm
[15:46:04] <brada> fuzzie: doesnt the string get messed up before setupString?
[15:46:30] <fuzzie> brada: SetupString is called from TextArea.
[15:46:38] <brada> ah that call
[15:47:04] <fuzzie> but i mean i don't see why it would happen, it should have no influence at all
[15:47:45] <brada> yeah
[15:47:50] <brada> and it doesnt for me
[15:48:09] <lynxlynxlynx> i didn't see a nice way to print whole tmp, but the last char was fine before exploding
[15:48:52] <brada> i dont know about gdb but i was just examining the memory directly
[15:51:15] <brada> did something get messed up for you because of all my rewriting history?
[15:52:18] <lynxlynxlynx> no
[15:52:46] <fuzzie> the PrintFromLine on line 88 *reads* from string directly which seems like it could potentially be weird
[15:53:11] <fuzzie> but .. only weird
[15:54:27] <fuzzie> vvery odd.
[15:55:37] <brada> clearly its an obscure bug since it doesnt manifest ofr me with the same data set
[15:55:46] <fuzzie> i mean, to me it looks like "[i\n" is the start of [initials which got broken up by SetupString, but that makes no sense at all since it's been nowhere near it
[15:57:31] --> Canageek has joined #gemrb
[15:58:40] <brada> this is maddening
[16:00:39] <lynxlynxlynx> at first i thought the length was wrong, 37 vs 39, but then i realised gdb did the " escaping
[16:02:40] <brada> well since you arent using multibyte you can assert that sterlen(str) == len
[16:02:50] <brada> but i doubt thats the issue
[16:03:27] <lynxlynxlynx> fuzzie: can you try reproducing?
[16:03:30] <fuzzie> not now :/
[16:03:52] <fuzzie> after dinner if need be (~2h)
[16:04:11] <lynxlynxlynx> probably traveler will be here by then :)
[16:04:32] <lynxlynxlynx> or some lurker here wakes up from hibernation
[16:05:19] <fuzzie> so far today I got no study and no work done, so it's not such a good trend for me being helpful at all
[16:05:31] <brada> where is chiv?!
[16:10:48] --> miha has joined #gemrb
[16:14:45] --> rocket_hamster has joined #gemrb
[16:17:54] <-- brada has left IRC (Quit: brada)
[16:25:22] <Seniorita> [commit] Jaka Kranjc: Interface::InitItemTypes: fixed the itemtypedata vector size http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=82305d583882968f3d4ad67b41d92006546f4ce6
[16:25:23] <Seniorita> [commit] Jaka Kranjc: GemRB.cpp: move the using directive, so also the old android fg/bg callbacks get it http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=22d2d132e77a78391ae6574ef8e2ab08a9206e8b
[16:42:56] <-- miha has left IRC (Ping timeout: 245 seconds)
[16:49:27] <-- rocket_hamster has left IRC (Read error: No route to host)
[16:50:03] --> brada has joined #gemrb
[17:40:58] --> miha has joined #gemrb
[17:44:27] --> fizzle has joined #gemrb
[17:46:46] <-- miha has left IRC (Ping timeout: 256 seconds)
[17:48:42] --> miha has joined #gemrb
[17:57:24] <-- miha has left IRC (Ping timeout: 248 seconds)
[17:59:28] --> miha has joined #gemrb
[18:04:57] <brada> well libsdl2 compiles for android no problem....
[18:08:34] <Seniorita> [commit] Jaka Kranjc: treat EA_EVILCUTOFF as nonhostile for saving purposes http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=689b3f10856637dc294430daf45b576c3237b265
[18:09:09] <lynxlynxlynx> i don't remember if there were any blockers left
[18:09:28] <lynxlynxlynx> i know initially we didn't have touch support due to different events, but that was fixed iirc
[18:24:23] <-- kida has left IRC (Ping timeout: 240 seconds)
[18:35:26] <-- Canageek has left IRC (Quit: KVIrc 4.0.4 Insomnia The future is here. It's just not widely distributed yet. - William Gibson)
[18:38:31] <Seniorita> [commit] Jaka Kranjc: GameScript::MarkSpellAndObject: avoid a potential crash http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=8de95412872362f7c7a4e08495b26a0196f51dc5
[18:38:32] <Seniorita> [commit] Jaka Kranjc: HitBy has a damage type second parameter in all games - set it http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=3af1237f8edabad8351bf80d8e448faf81e5b088
[18:38:33] <Seniorita> [commit] Jaka Kranjc: GameScript::UseItem: also consider undroppable items http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=ce537ed9fafb8b147f62fd6500d7e97043592037
[18:40:04] <brada> i have SOs for all required android dependancies if anybody needs them
[18:40:43] <fuzzie> for which arch? :P
[18:41:11] <fuzzie> would be nice to have prepackaged arm+armv7a ones if you built with gcc 4.6, I guess
[18:41:27] <fuzzie> but only once you confirmed you can actually make it work :)
[18:42:10] <fuzzie> if we can somehow get away with just building gemrb itself then a buildbot slave would be trivial (I could put one up)
[18:42:35] <brada> i already volunteered my bot if neeeded
[18:42:41] <brada> but just arm
[18:42:42] <brada> not v7
[18:42:52] <brada> im sure i can compile v7
[18:42:59] <fuzzie> yeah I was just wondering if it was arm
[18:43:23] <brada> compiling gemrb still eludes me atm
[18:43:27] <fuzzie> people often target x86 for initial release since obviously that's the only way you can get a usable build to simulate
[18:45:26] <fuzzie> how did you build libsdl though?
[18:45:40] <brada> just out of my SDL repo clone
[18:45:43] <fuzzie> but how?
[18:45:45] <brada> ndk-build
[18:46:14] <fuzzie> the jni/SDL/ one?
[18:46:35] <brada> SDL/android-project
[18:46:58] <fuzzie> I'm just guessing it's actually useless to distribute it.
[18:47:06] <brada> maybe
[18:47:08] <brada> why?
[18:47:15] <fuzzie> Since to build an app you're going to need the SDL tree anyway.
[18:48:05] <brada> i thought we would use the sdl supplied java activity
[18:48:14] <fuzzie> yes
[18:48:31] <fuzzie> but to get a package you still need all these bits at app build time, right?
[18:48:40] <brada> yes
[18:49:43] <brada> i dont know why none of the commandergenius stuff will compile for me :(
[18:49:56] <fuzzie> it's probably only tested on linux
[18:50:00] <brada> but for gemrb id rather develop our own build scripts
[18:50:00] <fuzzie> so it could be anything
[18:50:22] <fuzzie> well, the magic of the commandergenius stuff is that their sdl port works, as previously discussed
[18:50:37] <brada> beholder had gemrb working with standard sdl2
[18:50:38] <fuzzie> if you don't want their sdl port then I'm pretty sure nothing in their tree is going to help
[18:50:46] <fuzzie> yes, but define "working" :P
[18:50:51] <miha> compile breaks at your code, not commandergenius or sdl code
[18:51:01] <miha> ndk doesnt like your cpp files
[18:51:17] <brada> fuzzie: working as in a minor glitch in the touch events that can be fixed
[18:51:27] <fuzzie> brada: but does it work on devices without opengl es 2.0 for example?
[18:51:34] <brada> oh that i dont know
[18:51:50] <fuzzie> i mean, it's the "doesn't work on most hardware" which has always been the problem with upstream
[18:51:59] <fuzzie> but that's hardly relevant for making sure it works
[18:52:07] <fuzzie> miha: it's not commandergenius code anyway :P
[18:52:10] <brada> well gemrb is so slow its useless to run on older hardware imo
[18:52:41] <fuzzie> miha: pelya's tree used to build gemrb fine, so it seems weird that it'd all go wrong now
[18:53:17] <brada> well at least im not the only one it doesnt work for
[18:53:34] <brada> maybe i should try an older revision of CG
[18:53:36] <fuzzie> brada: well I think really it's silly to try gemrb on anything except modern tablets anyway.
[18:53:45] <fuzzie> but please stop calling it CG, you will break my brain
[18:53:53] <brada> but that is moot if we are to package our own build script
[18:53:57] <fuzzie> CG is a Keen engine. :P
[18:54:04] <brada> :D
[18:54:20] <miha> fuzzie: you always had to use some patch
[18:54:22] <lynxlynxlynx> ah, found why one other dialog range problem happens. In this area tree logs on the ground are doors
[18:54:28] <brada> so my point is why do we care if it works without opengl es 2?
[18:54:36] <lynxlynxlynx> just another case of the changed door bits :)
[18:55:38] <lynxlynxlynx> brada: you and miha have different compiling problems
[18:56:05] <fuzzie> I thought lynx patched the only remaining likely compile problem now.
[18:56:19] <fuzzie> Well, except pelya's tree doesn't exclude the right files, but that's obvious enough.
[18:56:43] <lynxlynxlynx> the odd thing is that i didn't fix the scope error for him
[18:56:47] <lynxlynxlynx> i don't get it
[18:57:10] <fuzzie> brada: I'm not sure we do, I'm just saying that the advantage of pelya's tree is wide hardware support, which we might care about at some point, I don't know.
[18:57:21] <lynxlynxlynx> alex's patch did it inside the android ifdef, but that way doesn't compile for anyone else
[18:57:30] <fuzzie> lynxlynxlynx: you have it before the Audio.h include
[18:57:42] <fuzzie> maybe that matteers somehow?
[18:58:16] <lynxlynxlynx> maybe, but he didn't respond after that
[18:58:21] <miha> about opengl
[18:58:21] <miha> Try to use GLES 2.x context - will revert to GLES 1.X if unsupported by device
[18:58:21] <miha> you need this option only if you're developing 3-d app (y) or (n) (n):
[18:58:36] <miha> isnt that opengl es? :)
[18:59:07] <fuzzie> miha: this is about upstream sdl, not pelya's tree
[18:59:45] <miha> hm
[18:59:46] <miha> ok
[19:00:03] <miha> btw, i reconfigured and it seems to compile now
[19:00:06] <fuzzie> cool
[19:00:13] <miha> will se how far i can get :)
[19:02:51] <miha> different error this time http://pastebin.ubuntu.com/1560440/
[19:02:53] <Seniorita> Ubuntu Pastebin
[19:03:02] <miha> but it was compiling some time
[19:03:29] <brada> i do know that (unless something has changed since we last tried) touch input wont work with his version of SDL 1.3
[19:03:44] <brada> and i doubt anything has changed since it is still refered to as SDL 1.3
[19:04:07] <fuzzie> yes, everyone is still using 1.2
[19:04:15] <brada> miha: you shouldnt be compiling that logger
[19:04:19] <brada> Win32ConsoleLogger
[19:04:24] <fuzzie> in fact the patch linked in the bug goes so far as to #ifdef out the SDL 2.0 gemrb driver
[19:05:02] <brada> yes, but we have compiled it with that
[19:05:06] <brada> beholder i mean
[19:05:22] <miha> brada: well i dont do c++ sorry
[19:05:29] <brada> me neither :p
[19:05:53] <miha> beyond helloworld++
[19:05:54] <miha> :P
[19:07:28] <lynxlynxlynx> that's odd, we have an explicit rule to only build the win32console for windows
[19:07:38] <fuzzie> this isn't using cmake
[19:07:44] <lynxlynxlynx> ooh
[19:07:44] <brada> yeah
[19:07:51] <brada> needs some ifdef i guess
[19:08:03] <lynxlynxlynx> then the one in the patch applies, meh
[19:08:11] <brada> since that build script indiscriminately compiles everything it finds
[19:08:47] <brada> btw miha: you should change the library linkage settings from SDL_ttf to freetype
[19:09:02] <lynxlynxlynx> https://gist.github.com/2016179/download#
[19:09:08] <lynxlynxlynx> just recreate the bit for this console
[19:09:22] <lynxlynxlynx> ifndef / endif
[19:10:03] <lynxlynxlynx> brada: do you have a link to the script source handy?
[19:13:31] <miha> i did, now fonts
[19:13:46] <brada> lynx: not sure what you are asking...
[19:14:02] <lynxlynxlynx> the hack that tries to build everything
[19:14:17] <lynxlynxlynx> miha: hmm, fonts are new, what's the error?
[19:14:31] <fuzzie> it's not a script as such
[19:14:42] <miha> jni/../jni/application/gemrb/gemrb/gemrb/gemrb/plugins/TTFImporter/TTFFontManager.cpp:69: error: class 'GemRB::TTFFontManager' does not have any field named 'ftStream'
[19:14:45] <miha> jni/../jni/application/gemrb/gemrb/gemrb/gemrb/plugins/TTFImporter/TTFFontManager.cpp:71: error: 'FT_Error' was not declared in this scope
[19:14:48] <fuzzie> well, not a gemrb-specific one.
[19:14:51] <miha> and few more in that file
[19:15:00] <brada> in the AndroidAppSettings.cfg file that commandergenius uses it just has AppSubdirsBuild='gemrb/gemrb/*'
[19:15:41] <brada> so it just recursively globs that directory for any c/cpp files
[19:16:12] <fuzzie> see the top of https://github.com/pelya/commandergenius/blob/sdl_android/project/jni/application/Android.mk for the delicious globbing
[19:16:14] <Seniorita> commandergenius/project/jni/application/Android.mk at sdl_android · pelya/commandergenius · GitHub
[19:16:26] <fuzzie> but it's a global makefile, not a gemrb one
[19:16:30] <brada> yeah
[19:16:32] <fuzzie> since as brad says it's just out of a cfg
[19:17:03] <brada> miha: you dont have to build the ttf importer
[19:17:12] <brada> it would be nice for us to fix that tho
[19:17:31] <lynxlynxlynx> yes
[19:17:37] <lynxlynxlynx> one step at a time though :)
[19:17:41] <miha> brada: well, fix if you can. patches are problem with every new version
[19:17:58] <miha> well
[19:18:06] <lynxlynxlynx> you have freetype installed?
[19:18:10] <fuzzie> it has to be fixed in pelya's tree I expect
[19:18:42] <brada> commandergenius has it yes
[19:18:46] <fuzzie> although if you did as brada said and replaced sdl_ttf with freetype in the cfg, I'm surprised it doesn't automagically work
[19:19:04] <brada> probably didnt do that yet :p
[19:20:18] <miha> lynxlynxlynx: libfreetype6 was installed, -dev is too now, doesnt change it
[19:20:33] <miha> does android ndk uses that? :)
[19:20:37] <fuzzie> nope
[19:20:52] <miha> fuzzie: i didnt
[19:20:54] <fuzzie> ndk doesn't provide libfreetype support
[19:20:59] <miha> fuzzie: i was looking for that
[19:21:00] <fuzzie> so you have to get it from pelya's tree
[19:21:01] <miha> in log
[19:21:34] <fuzzie> 20:08 <brada> btw miha: you should change the library linkage settings from SDL_ttf to freetype
[19:21:43] <fuzzie> :-)
[19:21:47] <miha> fuzzie: where?
[19:21:55] <fuzzie> in the cfg file in pelya's tree
[19:22:07] <fuzzie> i.e. the AndroidAppSettings.cfg
[19:22:55] <fuzzie> I *assume* the include paths should get dealt with from that. I don't see anywhere else.
[19:23:04] <brada> yes
[19:23:53] <miha> yes
[19:24:00] <fuzzie> works?
[19:24:58] <Seniorita> [commit] Jaka Kranjc: Door::UpdateDoor: handle the iwd2 transparent bit too http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=ad8000f0c3ac62ef7e7ccdf6c5439fdc16c486ae
[19:25:40] <miha> SDL2video error
[19:25:40] <miha> http://pastebin.ubuntu.com/1560511/
[19:25:42] <Seniorita> Ubuntu Pastebin
[19:25:46] <brada> yeah thats expected
[19:25:52] <brada> you cannot compile sdl2 code
[19:25:56] <brada> and sdl1 code
[19:26:12] <fuzzie> it gives a pretty good try though, that's not much which fails :P
[19:26:13] <brada> that patch just ifdeffed out the entire sdl20video.cpp file
[19:26:32] <brada> so do that for now
[19:27:47] <brada> fuzzie: more than that fails im sure the compiler jsut shuts up after so many errors
[19:27:59] <brada> SDL_Texture and SDL_renderer come to mind
[19:29:42] <fuzzie> I wouldn't be surprised if pelya's SDL 1.2 has all that.
[19:29:58] <brada> o_O
[19:29:59] <fuzzie> Since it's only complaining about a missing member of the touch event struct, not the struct not existing..
[19:30:19] <miha> now vlc.h problem (not in patch) is that new?
[19:30:31] <fuzzie> that you should also disable I guess
[19:30:36] <brada> you wont be able to compile that
[19:30:37] <fuzzie> what *is* the VLC plugin for? :P
[19:30:51] <brada> so i can watch the mac versions quicktime movs
[19:30:53] <brada> :p
[19:30:56] <fuzzie> ah
[19:30:57] <fuzzie> lazy :P
[19:30:59] <brada> heh
[19:31:07] <brada> and it was a learning exercise
[19:31:26] <brada> and it should work with BGEE ;)
[19:31:27] <fuzzie> implementing quicktime is far more fun
[19:31:34] <brada> *shudder*
[19:31:39] <fuzzie> trust me on this
[19:32:04] <fuzzie> (obviously your idea of fun may differ from torture)
[19:32:15] <brada> im not much of a masochist...
[19:33:45] <fuzzie> of course vlc doesn't implement any of the quicktime codecs I've needed anyway, so I didn't get to use that escape route
[19:33:50] <brada> actually i may have to recant that statement after all this font business :p
[19:34:20] <fuzzie> hehe
[19:34:22] <fuzzie> yes :)
[19:34:36] <brada> i think maybe vlc on mac will hook into the quicktime framework cuz it has played every quicktime movie i tried
[19:35:17] <fuzzie> the trouble is
[19:35:18] <brada> even the ones that quicktime no longer plays :p
[19:35:32] <fuzzie> quicktime allowed old games to add their own codecs
[19:35:41] <brada> ew
[19:35:43] <fuzzie> and oddly enough modern quicktime doesn't like codecs which are only provided as m68k binaries
[19:35:51] <brada> ha ha ha
[19:35:52] <fuzzie> i can't imagine why. super laziness on apple's part.
[19:36:13] <brada> quicktime was so terrible
[19:36:23] <fuzzie> but quicktime had all kinds of insane stuff anyway
[19:36:29] <brada> exactly
[19:36:34] <fuzzie> quicktime vr, quicktime sprites, flash movies inside quicktime
[19:36:42] <fuzzie> flash inside quicktime inside flash inside quicktime inside flash inside quicktime
[19:36:45] <fuzzie> and so forth
[19:36:59] <brada> yes its quite nauseating
[19:37:14] <brada> tho now its not so bad
[19:37:25] <fuzzie> yes, as long as you don't want to play old video, ever :P
[19:37:27] <miha> error sdl1.3 video keymap http://pastebin.ubuntu.com/1560546/
[19:37:27] <Seniorita> Ubuntu Pastebin
[19:37:28] <brada> probably why not all quicktime movies will play in qt anymore
[19:38:03] <brada> miha: there should be a patch file for that
[19:38:27] <brada> in the application/gemrb-1.3 directory
[19:39:09] <brada> hmm maybe not what i was thinking of
[19:39:30] <brada> i see a bunch of #define SDLK_KP* stuff
[19:40:00] --> Yoshimo has joined #gemrb
[19:40:11] <brada> i thought you were building with sdl 1.2
[19:41:01] <miha> i was at first, then i tried 1.3
[19:41:02] <miha> so
[19:41:11] <miha> should i build with 1.2 ? :)
[19:41:27] <miha> nobody told me to
[19:41:40] <brada> yes try 1.2
[19:41:55] <fuzzie> ah that explains the 2.0 build almsot working :)
[19:42:08] <brada> yes it does
[19:43:41] <brada> this font bug would probably be fixed a lot faster if i merged and broke it for everybody :p
[19:43:48] <-- Yoshimo has left IRC (Client Quit)
[19:44:54] <miha> well with 1.2 i get jni/../jni/application/gemrb/gemrb/gemrb/gemrb/GemRB.cpp:45: error: 'void GemRB::appPutToBackground()' should have been declared inside 'GemRB'
[19:44:58] <miha> my first error
[19:45:03] <miha> that's why i tried 1.3 :p
[19:45:55] <miha> lynxlynxlynx told me to try something but i didnt know how and then internet connection dropped
[19:47:17] <brada> hiha: you dont get that with 1.3 because there is an ifdef stopping it with 1.3
[19:47:21] <brada> miha
[19:47:54] <miha> brada: yes i saw that
[19:47:58] <miha> but what should i do?
[19:48:20] <lynxlynxlynx> try moving it after the includes
[19:48:23] <brada> well where is your namespace GemRB at?
[19:48:46] <brada> tho why is that needed?
[19:49:09] <brada> they are declared as GemRB::*
[19:50:12] <fuzzie> they shouldn't be
[19:50:16] <fuzzie> get rid of the GemRB::
[19:50:49] <brada> you can probably get rid of those functions all together
[19:50:54] <fuzzie> can you?
[19:50:59] <brada> that stuff should be handled regardless of sdl version
[19:51:14] <fuzzie> oh?
[19:51:17] <brada> iirc
[19:51:22] <miha> fuzzie: just remove that? :)
[19:51:22] <brada> but you know me an iirc :p
[19:51:26] <fuzzie> miha: yes
[19:52:15] <brada> well i guess they arent handled in sdl 1.2
[19:52:18] <brada> but they can be
[19:52:20] <fuzzie> i don't think they're handled at all
[19:52:32] <fuzzie> how would that work?
[19:53:35] <brada> im trying to remember
[19:53:49] <miha> compiling...
[19:53:51] <fuzzie> for the main thread, the default for both pelya's and official is to block at SDL_Flip
[19:54:05] <fuzzie> but that doesn't do anything about other threads
[19:55:30] <fuzzie> for SDL 2.0 you can handle this with SDL_WINDOWEVENT_MINIMIZED/RESTORED, or FOCUS_LOST/GAINED
[19:55:41] <brada> i think 1.2 can use SDL_ActiveEvent
[19:55:46] <fuzzie> but I think you're stuck with some kind of hack in 1.2
[19:56:31] <brada> http://www.libsdl.org/cgi/docwiki.cgi/SDL_ActiveEvent
[19:56:35] <Seniorita> SDL_ActiveEvent - SDL Documentation Wiki
[19:56:36] <fuzzie> yes, but does anything do it?
[19:57:03] <miha> linking problem http://pastebin.ubuntu.com/1560589/
[19:57:04] <Seniorita> Ubuntu Pastebin
[19:57:07] <brada> you mean does sdl 1.2 on android fire that?
[19:57:12] <fuzzie> brada: yes
[19:57:34] <brada> dont know
[19:58:01] <fuzzie> miha: that's odd, that call is deliberately wrapped in an SDL 1.3+ check
[19:58:14] <brada> yes yes it is...
[19:58:24] <fuzzie> maybe you need to remove that SDLVideo.o file and let it be rebuilt
[19:58:30] <brada> yes probably
[20:00:42] <miha> where is the .o? :$
[20:00:56] <fuzzie> you pastebinned the path
[20:01:09] <brada> ./obj/local/armeabi/objs-debug/application/gemrb/gemrb/gemrb/gemrb/plugins/SDLVideo/SDLVideo.o
[20:01:21] <miha> i dont have it
[20:01:46] <brada> you must have it
[20:02:13] <miha> if ifndef it before? :)
[20:02:16] <miha> ok
[20:02:27] <fuzzie> well I guess you can just randomly change the .cpp file :P
[20:03:04] <brada> oh yeah
[20:03:07] <brada> that ought to work
[20:06:42] <miha> http://pastebin.ubuntu.com/1560611/
[20:06:42] <Seniorita> Ubuntu Pastebin
[20:06:50] <miha> lots of lines this time
[20:07:41] <brada> yes because you are trying to compile sdl20video.cpp
[20:07:48] <brada> you need to ifdef out that entire file
[20:08:09] <miha> but then there is no SDLVideo.o ?
[20:08:19] <brada> sure there will be
[20:08:31] <brada> you still comiple sdl12video.coo
[20:08:34] <brada> and sdlvideo.cpp
[20:08:39] <brada> just not sdl20video
[20:11:25] <miha> ok
[20:11:25] <miha> /home/miha/geminstall/commandergenius/project/jni/../jni/application/gemrb/gemrb/gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:545: undefined reference to `SDL_GetColorKey'
[20:11:45] <miha> ld error..
[20:12:58] <brada> yeah thats why we asked you to clean that o file
[20:13:17] <brada> since that condition shouldnt be hit when compiling for SDL < 1.3
[20:13:39] <fuzzie> but making any change to the SDLVideo.cpp (e.g. adding a space) should force a recompile too.
[20:14:00] <brada> if you just make an edit to SDLVideo.cpp then undo and save the file it should do it automagically
[20:14:05] <brada> what fuzzie said
[20:15:25] <brada> just make sure you notice we are talking about SDLVideo.cpp and not either of the variants
[20:15:50] <miha> thank you for your patience :$
[20:16:00] <brada> no problem
[20:16:04] <brada> thank you for doing this
[20:16:16] <brada> we have needed a new android build for months
[20:17:31] <miha> -rw-rw-r-- 1 miha miha 6114991 jan 22 21:16 MainActivity-debug.apk
[20:17:46] <miha> this is it?
[20:17:55] <fuzzie> that looks about right
[20:17:59] <fuzzie> can you try it? :)
[20:18:34] <fuzzie> presumably it's still pretty annoying to play without brad's touch stuff
[20:18:46] <miha> i dont even know how to play :)
[20:18:53] <miha> will see if it crashes though
[20:21:14] <miha> offsers me game download.. demo
[20:21:28] <brada> yeah so it seems to work then
[20:21:51] <brada> of course you still need to go though the nightmare of making a gemrb.cfg if you actually wish to play
[20:22:13] <miha> http://www.wolfey.si/tmp/gemrb.apk
[20:24:24] <miha> i chose the other option.. and it counted 2 things to 100% and closed
[20:24:28] <miha> no error though
[20:25:36] <brada> fuzzie:
[20:25:37] <brada> http://paste.debian.net/227569/
[20:25:38] <Seniorita> debian Pastezone
[20:25:54] <brada> snippit from some game in commandergenius
[20:26:03] <brada> so it looks like that should work
[20:26:54] <brada> miha: i dont know what option you speak of
[20:26:58] <fuzzie> brada: which game?
[20:27:05] <brada> the interface you see there has nothing to do with gemrb
[20:27:30] <brada> enigma
[20:27:35] <fuzzie> thx
[20:27:38] <miha> brada: instead of downloading baldurs gate demo, i chose option to overwrite something
[20:27:51] <brada> yeah like i said that is out of our hands
[20:28:13] <fuzzie> you should download demo if you don't want to setup gemrb.cfg
[20:28:48] <fuzzie> brada: so, my concern would be, what's stopping the main thread from being blocked before you get that message?
[20:28:57] <brada> oh i see
[20:29:22] <miha> will try tomorrow, here i have very bad wifi connection and i dont want to pay for 800 MB gprs :)
[20:29:48] <fuzzie> brada: I only just thought of that now, I don't mean to imply I thought that all along!
[20:29:58] <miha> thx again for your patience, it was very.. informative
[20:30:02] <brada> it blocks in SDL_Flip?
[20:30:09] <fuzzie> i believe so, yes
[20:30:55] <lynxlynxlynx> miha cool
[20:32:34] <miha> lynxlynxlynx: i just wonder why teach me to build since you guys told me everything. but thx :)
[20:32:35] <lynxlynxlynx> i've ran through the whole sheangarne
[20:32:53] <lynxlynxlynx> apart from some spawn problems and a tiny script one, everything is fine
[20:33:04] <lynxlynxlynx> miha: you have the toolchain set up
[20:33:42] <miha> not much toolchain, android sdk and ndk and ant.
[20:33:57] <lynxlynxlynx> more than me
[20:34:00] <miha> :)
[20:34:15] <lynxlynxlynx> and in this time, i fixed three bugs, so delegation works :)ž
[20:34:25] <fizzle> hm, is totsc broken with gemrb?
[20:34:42] <fizzle> I can't seem to leave Ulgoth's Beard...
[20:34:59] <lynxlynxlynx> parts of durlag's tower, but everything should be well documented on the todo
[20:35:50] <lynxlynxlynx> previous reports were either not complete or we improved enough to break things again (that's why totsc was said to work a bunch of releases ago)
[20:36:43] <fizzle> how is the worldmap update supposed to work?
[20:36:53] <fizzle> (because that seems not to have happened)
[20:37:11] <fizzle> I don't get UB or DT on the map
[20:37:25] <fizzle> so I can't leave UB because distance calculation fails
[20:39:35] <brada> then how did you get to UB in the first place? o_O
[20:40:42] <fizzle> automatic teleport (from a save where I was right before the final duel)
[20:41:14] <fizzle> I thought that was only supposed to happen when you had actually beaten the game already but maybe not
[20:42:22] <brada> no you can go to both places before you beat the game
[20:43:00] <brada> i remember using cheap backstab tactics to do DT at a low level an then being over powered the rest of the game :D
[20:44:17] <fizzle> yes, but the teleport I think should only happen when you install totsc after having finished bg
[20:44:18] <lynxlynxlynx> yeah, but you have to talk to someone to start it
[20:44:32] <lynxlynxlynx> auto nicking from the final battle is not good
[20:44:49] <lynxlynxlynx> i don't know if you can exit ub via the worldmap
[20:44:57] <lynxlynxlynx> but i never played totsc
[20:45:09] <lynxlynxlynx> being an island and all ...
[20:46:19] <fizzle> Ub is no island
[20:47:32] <fizzle> ah, the teleport is in the Undercity area script
[20:47:50] <fizzle> so I guess that was okay
[20:49:24] <fizzle> does the worldmap get saved in the savegame?
[20:49:29] <fuzzie> yep
[20:49:57] <fizzle> so the game is using the old worldmap data from before installing the exp probably?
[20:50:53] <fuzzie> hm
[20:52:05] <fuzzie> quite possible
[20:52:31] <fuzzie> we only handle upgrading games for bg2, iwd and iwd2, it looks like
[20:52:41] <fuzzie> ('GameExpansion' function in guiscript)
[20:53:05] <fuzzie> and iwd/iwd2 ones are just hacks anyway
[21:00:39] <brada> fuzzie: have time to try mbfonts branch?
[21:40:53] <-- miha has left IRC (Quit: Lost terminal)
[21:59:17] --> traveler__ has joined #gemrb
[22:02:43] --> edheldil_ has joined #gemrb
[22:34:10] <brada> found a bit of a problem in the font code. dont think it will help lynx
[22:35:02] <brada> but maybe
[22:38:02] <brada> lynx: you can try the latest of that branch
[22:38:20] <brada> i doubt that will fix your problem since you were using english...
[22:40:03] <lynxlynxlynx> tommorow
[22:42:16] <brada> fuzzie: i have confirmed now that state glyphs > 127 work
[22:59:59] <-- brada has left IRC (Quit: brada)
[23:05:48] <fizzle> does this look sensible: http://paste.debian.net/hidden/b70699fb/
[23:05:49] <Seniorita> Debian Pastezone
[23:15:39] <-- fizzle has left #gemrb
[23:20:33] <lynxlynxlynx> not sure all of that is needed
[23:20:44] <lynxlynxlynx> chapter 1 of iwd2 is also very playable :)
[23:21:32] <lynxlynxlynx> fuzzie: looks like all those split action/dialog ids files and scripting differences don't play that much of a role :)
[23:24:38] <-- traveler__ has left IRC (Ping timeout: 245 seconds)
[23:28:45] --> nutron has joined #gemrb
[23:36:46] <lynxlynxlynx> note to self: Scripts: <none> <none> <none> <none> <none> 3101brsc 00adgpmr 3101brsm
[23:36:49] <lynxlynxlynx> good night
[23:37:31] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:37:48] --> traveler__ has joined #gemrb