#gemrb@irc.freenode.net logs for 17 Sep 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:04:47] --> brada has joined #gemrb
[00:24:34] --> edheldil_ has joined #gemrb
[00:41:30] <-- Earnestly has left IRC (Quit: WeeChat 0.3.9-dev)
[01:41:38] <-- edheldil_ has left IRC (Ping timeout: 240 seconds)
[02:28:38] <-- wrotek__ has left IRC (Ping timeout: 240 seconds)
[03:33:49] <-- brada has left IRC (Quit: brada)
[03:41:17] --> brada has joined #gemrb
[05:00:59] --> wrotek__ has joined #gemrb
[05:26:02] --> decker_ has joined #gemrb
[05:26:32] <decker_> greetz
[05:26:37] <decker_> haven't been here in forever
[05:44:39] <-- decker_ has left IRC (Quit: Page closed)
[06:03:59] <-- Textmode has left IRC (Ping timeout: 256 seconds)
[06:04:52] --> edheldil_ has joined #gemrb
[06:08:23] <-- brada has left IRC (Quit: brada)
[06:17:55] <-- edheldil_ has left IRC (Ping timeout: 246 seconds)
[06:19:18] --> Textmode has joined #gemrb
[06:24:14] <-- Textmode has left IRC (Ping timeout: 255 seconds)
[06:38:35] --> Textmode has joined #gemrb
[06:58:27] <-- Textmode has left IRC (Ping timeout: 255 seconds)
[07:29:37] --> edheldil has joined #gemrb
[07:29:37] --- ChanServ gives channel operator status to edheldil
[07:38:34] --> Textmode has joined #gemrb
[07:43:01] <-- Textmode has left IRC (Ping timeout: 255 seconds)
[07:53:36] --> lynxlynxlynx has joined #gemrb
[07:53:36] <-- lynxlynxlynx has left IRC (Changing host)
[07:53:36] --> lynxlynxlynx has joined #gemrb
[07:53:36] --- ChanServ gives channel operator status to lynxlynxlynx
[08:13:48] <lynxlynxlynx> got a reply from the head of obsidian
[08:14:19] <lynxlynxlynx> he already discovered gemrb half a year ago; will forward to i guess a more technical person
[08:31:25] <edheldil> interesting ...
[08:32:25] --> Husar1 has joined #gemrb
[08:33:10] --> Avenger_teambg has joined #gemrb
[08:33:27] <Avenger_teambg> lynxlynxlynx thats interesting :)
[08:34:05] <Avenger_teambg> i mean, if it goes in the technical direction, it can only be good
[08:34:12] <lynxlynxlynx> i guess chris told him about it, since i remember his mention on twitter
[08:38:52] <Avenger_teambg> if it goes commercial, some kind of drm support is needed. no one wants to give their artistic content for free
[08:43:32] <lynxlynxlynx> they actually have a "no drm" tier
[08:43:38] <lynxlynxlynx> pretty long down the road though
[08:45:27] <Avenger_teambg> uhm, which is first linux or no drm :D
[08:45:56] <Avenger_teambg> because, if they don't plan linux, i won't even talk to them :D
[08:46:33] <fuzzie> linux is at $2.2m or so?
[08:46:39] <fuzzie> yes, $2.2m
[08:47:04] <Avenger_teambg> and no drm?
[08:47:50] <fuzzie> looks like 'coming soon' to me
[08:48:50] <fuzzie> am pretty sure that the 'we need drm to protect our content' argument has been thoroughly discredited though :P
[08:51:48] <Avenger_teambg> well, some kind of drm definitely protects revenue. It is just not linear correlation
[08:52:07] <Avenger_teambg> if you put it on the net as free download, you will get little revenue :D
[08:52:47] <Avenger_teambg> on the other hand, a very tough drm effectively bars some rightful owners and it is contra productive
[08:53:59] <fuzzie> well, the problem is, weak DRM gets broken in a matter of hours, so it goes on the bittorrent sites etc just as quickly, so it doesn't really help
[08:54:17] <Avenger_teambg> tough drm gets broken the same time too
[08:54:27] <Avenger_teambg> so, the weak drm wins
[08:54:34] <fuzzie> well, some really tough DRM has taken days to break
[08:54:52] <fuzzie> but of course it usually really annoys a lot of customers
[08:55:11] <fuzzie> but at least then, you have some days in which people might go to the torrent site, not find a working crack, and go and buy it instead
[08:55:22] <Avenger_teambg> yeah well, i guess it didn't worth the effort, counting the annoyed customers whose drives broke :D
[08:56:05] <fuzzie> well they already paid for the product, so sucks to be them!
[08:56:32] <fuzzie> pretty sure that obsidian aren't idiots though, honestly.
[08:56:37] <Avenger_teambg> anyway, i just want to figure out how could we have gemrb and contemporary + proprietary dataset
[08:57:29] <fuzzie> well I think realistically they'll already have their own engine and it'll be completely new and probably using all kinds of 3D effects etc
[08:58:07] <Avenger_teambg> i doubt, if they just gather money for it
[08:58:57] <fuzzie> and in any case their lawyers will probably want to stay way away from anything which could possibly be bioware's stuff
[08:59:05] <Avenger_teambg> ok, actually, you could be right, there is more than just the engine.
[08:59:10] <fuzzie> but it would be way cool if they acknowledged we exist and were friendly to the idea of open-source stuff
[09:00:10] <fuzzie> since then maybe we could avoid needing 10 years of reversing for whatever they do :P
[09:03:18] --> Avenger_teambg_ has joined #gemrb
[09:03:20] <-- Avenger_teambg has left IRC (Ping timeout: 245 seconds)
[09:03:36] <lynxlynxlynx> i explicitly mentioned we aren't completely decoupled from dnd yet
[09:04:21] <lynxlynxlynx> it's not hard to finish it if you don't care about generalisation
[09:04:25] --- Avenger_teambg_ is now known as Avenger_teambg
[09:04:36] <Avenger_teambg> Yeah, that would be a good exercise
[09:05:01] <Avenger_teambg> a totally new opcode pack wouldn't be bad either
[09:05:15] <Avenger_teambg> and we could also make the scripting engine a plugin
[09:05:33] <Avenger_teambg> that would really cut some dnd and even IE ties
[09:15:24] <edheldil> well, that's something I always wanted
[09:15:43] <edheldil> lynxlynxlynx> it's not hard to finish it if you don't care about generalisation <-- finish what? Decoupling?
[09:16:04] <lynxlynxlynx> yes
[09:16:36] <-- Avenger_teambg has left IRC (Quit: Page closed)
[09:17:14] <edheldil> maybe we should put pluginizing gamescript to gsoc ideas page
[10:03:29] --> Earnestly has joined #gemrb
[10:10:33] <lynxlynxlynx> sure, go ahead
[10:10:50] <lynxlynxlynx> it's a nice list of a-bit-larger-but-managable ideas
[11:51:21] <edheldil> lynxlynxlynx: ok, it's there: http://www.gemrb.org/wiki/doku.php?id=developers:gsoc_ideas
[11:52:15] <wjp> if it's really intended as a gsoc idea, it should involve more than just refactoring though, IMO
[11:52:52] <edheldil> implement a sample ruleset
[11:53:54] <edheldil> "gsoc" there might be a bit extraneous
[11:58:57] <Earnestly> "Create a GUI inside GemRB that would allow it to start without a config file, àla ScummVM." -> Please don't
[11:59:13] <Earnestly> Unless it's for android or something embeded
[11:59:57] <fuzzie> that is the general idea.
[12:00:39] <fuzzie> brad ended up doing a custom frontend for iOS I think.
[12:02:29] <Earnestly> That's fine, but it seems totally unnecessary. Conceptually, is it not feasable to just have an icon on iOS (I love the eytmology of that word) or Android linking it to `gemrb -c foogame' instead?
[12:03:27] <fuzzie> but then how would you do the configuration?
[12:04:01] <Earnestly> For that, some frontend like retroarch's GUI?
[12:04:18] <Earnestly> It's basically just a GUI abstraction of the config file
[12:04:25] <fuzzie> on android and iOS and etc?
[12:04:52] <Earnestly> Yeah.
[12:05:26] <Earnestly> Do you mean retroarch itself? That runs on almost everything tmk. But the concept seems easy enough
[12:06:08] <Earnestly> But, who's to say. If someone feels like making the effort then it's fair enough.
[12:06:34] <Earnestly> And writing new code is probably more fun than fixing existing :-P
[12:06:41] <fuzzie> well, I'm just wondering why you think it'd be viable
[12:06:58] <Earnestly> Seems transparent enough
[12:07:02] <fuzzie> retroarch doesn't seem to have much android code, but the little code it does have consists of a native file chooser
[12:07:31] <Earnestly> It's a frontend library for different emulators, I don't use it on a phone
[12:07:45] <fuzzie> well, I mean, we *don't* want to make the effort, is the thing
[12:07:57] <fuzzie> so someone can write a one-time built-in GUI which does the config
[12:08:08] <fuzzie> or else we can repeat some custom frontend for every single platform
[12:08:31] <Earnestly> Yes, I think we're both talking about the same thing here, so fair enough :p
[12:09:09] <Earnestly> "one-time built-in GUI which does the config", yes
[12:09:10] <fuzzie> (and you can't run binaries like that on android, no)
[12:09:26] <Earnestly> I guess this confused me " Create a GUI inside GemRB that would allow it to start without a config file"
[12:09:45] <fuzzie> the use case is "user installs GemRB onto their phone, puts the game content on their SD card, runs GemRB"
[12:10:13] <Earnestly> Btw, on a bit of a tangent, how do you play IE games on a phone?
[12:10:22] <fuzzie> well I personally don't :)
[12:10:50] <Earnestly> Ditto
[12:11:24] <fuzzie> would be nice on tablet but will have to wait for bgee improvements for the GUI to be usable, I guess
[12:12:38] <Earnestly> What on earth is the story behind bgee? I've heard about it but never fully investigated. Any relevant links you can share while I google-fu?
[12:13:05] <fuzzie> well, all I know is that they licensed the engine+content and are happily improving it
[12:13:10] <fuzzie> others here know more I'm sure :P
[12:13:52] <Earnestly> Still closed engine?
[12:14:11] <fuzzie> very much so
[12:14:38] <fuzzie> they're happy to prove bits of info though, and they're using Avenger's open-source editing tool to do the work
[12:15:27] <fuzzie> um, not prove. share?
[12:15:49] <Earnestly> Hm, they advertise Android and OSX, reckon Linux might creep in?
[12:16:38] <Earnestly> And how does, if at all, this relate to any of the other IE games?
[12:16:51] <fuzzie> Really I doubt it, but it's no more obscure than Android tablets are.
[12:17:24] <fuzzie> The rights to IWD and in particular PST are held by Interplay, so I hear they have no chance of getting them, but you never know.
[12:17:52] <Earnestly> Hm, the whole situation is somewhat depressing. But it's the way things are
[12:18:23] <Earnestly> On a more positive note, seems the same team have remade MDK2 as well.
[12:19:56] <Earnestly> fuzzie: Let's hope Interplay (they still exist?) takes a page or two from id software and simply release the older engines as open source.
[12:20:53] <fuzzie> they only own partial rights here
[12:22:57] <Earnestly> At least gemrb is here to slow down the bitrot in closed source code
[12:30:16] <edheldil> Earnestly: and some rights has also Hasbro, so ...
[12:56:23] <Earnestly> edheldil: Is Hasbro particularly bad when it comes to rights?
[12:58:23] <edheldil> nothing off the top of my head, but I think so
[13:01:06] <Earnestly> It can't be helped, hopefully I can contribute to open source some day
[13:14:38] <lynxlynxlynx> you can already today
[13:14:48] <lynxlynxlynx> it's a silly misconception unless you're out of time
[13:15:17] <-- kida_laptop has left IRC (Read error: No route to host)
[13:16:28] --> kida_laptop has joined #gemrb
[13:31:35] * edheldil raises an 'ACT NOW!' poster
[14:11:08] <Earnestly> lynxlynxlynx: Yup, I do contribute where I can. Mostly documentation
[14:11:56] <lynxlynxlynx> see, no need to talk about it like it will only happen sometime in the future
[14:18:50] <Earnestly> It's a personal goal of mine, to understand code more in depth. It's quite frustrating that I haven't really gone into depth with it
[14:20:45] <edheldil> According to one movie I saw, the best thing to enhance your coding skillz is to hold a loaded gun to your head ;-)
[14:21:09] <edheldil> I am sure there was more to it, but I aaah can't remember :-)
[14:21:26] <Earnestly> Haha what
[14:23:34] <lynxlynxlynx> the last occurrence i remember was in operation swordfish
[14:31:45] <edheldil> exactly :-)
[15:04:11] <-- CIA-45 has left IRC (Ping timeout: 245 seconds)
[15:17:35] --> brada has joined #gemrb
[15:40:08] --> CIA-40 has joined #gemrb
[15:40:33] <brada> using gemrb as a commercial game engine would be more practical if we werent stuck using SDL 1.2
[15:46:27] <Earnestly> brada: See the top of this list http://www.gemrb.org/wiki/doku.php?id=developers:gsoc_ideas
[15:47:20] --> Avenger has joined #gemrb
[15:47:20] --- ChanServ gives channel operator status to Avenger
[15:47:26] <Avenger> heh brada, you can always add a new video plugin
[15:47:39] <brada> i did start the sdl 2 plugin
[15:47:56] <Avenger> but then how we are stuck :D
[15:47:56] <brada> but i need a wjp or somebody else to help
[15:48:06] <Avenger> a fuzzie?
[15:48:11] <brada> that would do :)
[15:49:10] <Avenger> what's the problem? and why sdl 1.2 isn't good enough?
[15:52:56] <brada> sdl 1.2 doesnt support any modern mobile systems
[15:53:22] <brada> the android "support" is by third party
[15:53:43] <brada> and it doesnt support multitouch anyway
[15:55:22] <Avenger> ahh, i see.
[15:55:44] <Avenger> so sdl 2 is needed 'only' for the mobile part
[15:55:54] <Avenger> and we still have missing parts?
[15:56:18] <brada> well needed i guess, but also 1.2 is depricated so its only a matter of time before it starts breaking on other platforms
[15:56:53] <Avenger> well, i wouldn't mind sdl 2 on pc
[15:57:56] <Earnestly> 1.2 is depricated?
[15:58:13] <Earnestly> Where do you see that?
[15:58:28] <brada> well im exagerating a bit
[15:58:36] <brada> they still patch it here and there
[15:58:43] <brada> but no neew features will be in it
[15:58:44] <Earnestly> Ah okay. Sort of like saying Python2.7 is depricated :P
[15:58:48] <brada> yes
[15:59:08] <Avenger> well, python 3 is dangerous
[15:59:17] <brada> why?
[15:59:19] <Avenger> that would probably break us
[15:59:28] <Earnestly> The unicode changes?
[15:59:47] <brada> probably part of it
[16:00:00] <Avenger> i don't know, something broke when we 'just tried to use' it
[16:00:10] <Earnestly> Have fun splitting a string of two chinese characters and ending up with one \o
[16:00:42] <brada> heh you say that like gemrb supports MB strings :p
[16:01:04] <Avenger> well, there are plans to support chinese/korean no?
[16:02:17] <-- Husar1 has left IRC (Quit: Husar1)
[16:03:56] <brada> plans yes. they have born no fruit yet
[16:04:48] <Earnestly> https://www.youtube.com/watch?v=sgHbC6udIqc
[16:05:03] <Earnestly> pycon 2012: Pragmatic Unicode, or, How do I stop the pain?
[16:06:05] <edheldil> there are of course places where it would have to be changed, but do not forget that python has unicode support for quite a long time
[16:06:46] * |Blaze| Avenger hey
[16:07:18] <edheldil> the difference is that in py3k, you have to state explicitly what is string and what's not; you can't just quickly move your hands and hope that nobody sees it :)
[16:09:39] <Avenger> hello Blaze
[16:12:05] <brada> any move to python 3 should wait till all possible guiscripts are merged anyway
[16:12:50] <|Blaze|> hey
[16:17:33] <-- brada has left IRC (Quit: brada)
[16:25:42] --> Yoshimo has joined #gemrb
[16:59:50] <lynxlynxlynx> py3 compatibility would mostly be done by the migrator script (print semantics)
[17:00:42] <lynxlynxlynx> but we get no gain from a move, only pain
[17:01:17] <lynxlynxlynx> while most of the platforms still ship py2 and some maybe only py2, it's too early to move
[17:22:57] <fuzzie> I don't think we'd gain anything from a move until python2 support starts to vanish.
[17:23:22] <fuzzie> And pushing it for unicode support while python 3.3 is *still not released* is a really silly argument. :P
[17:24:08] <Avenger> fuzzie what about sdl 2?
[17:24:34] <fuzzie> well, I don't see the point of moving unless you want to use SDL 2 properly
[17:25:00] <Avenger> well, if properly means speed up :D
[17:25:14] <fuzzie> well, it means 'rewrite to use textures'
[17:25:26] <fuzzie> so, sure, it would be a speedup :P
[17:25:34] <lynxlynxlynx> can they be used also on glless machines?
[17:25:52] <lynxlynxlynx> ie, is it just a sdl structure or do you mean the usual texture
[17:26:02] <fuzzie> but SDL 2 really isn't optimal for software rendering
[17:26:23] <lynxlynxlynx> we already have it split up, so sdl1 can stay as the fallback
[17:26:47] <fuzzie> but yes, it would be nice to have the time to do it properly
[17:29:15] <fuzzie> although possibly just doing OpenGL would still be easier
[17:29:25] <Avenger> a 3rd plugin?
[17:30:55] <fuzzie> SDL2's API probably isn't quite broad enough to do all the fancy character effects.
[17:33:02] <fuzzie> hm, I guess you can just use SDL_GL_BindTexture and call OpenGL directly from SDL2 code
[17:33:48] <Avenger> what level of opengl is that?
[17:34:05] <Avenger> i mean, if you just put it in sdl2, will that break anything?
[17:35:15] <fuzzie> well, if you keep SDL 1.2 for software rendering, no
[17:35:27] <fuzzie> really I don't like to guess what SDL 2 upstream do though
[17:35:34] <Avenger> hehe
[17:36:23] <fuzzie> and whenever I talk to brada he doesn't seem any more confident than I am
[17:36:26] <Avenger> i would suggest you create a separate plugin, like 'fuzzie's opengl plugin' so you and brada won't collide.
[17:36:55] <lynxlynxlynx> it was more chaotic before, since there wasn't any sdl2, just a sdl1.3
[17:37:02] <lynxlynxlynx> now a release is actually in sight
[17:37:14] <Avenger> a new release? we barely did anything
[17:37:25] <lynxlynxlynx> of sdl
[17:37:31] <Avenger> aah, :D
[17:37:42] <lynxlynxlynx> our new one will happen in the first half of october
[17:38:13] <Avenger> we should kill some of the pending bugs
[17:38:21] <lynxlynxlynx> don't forget we left that active-by-default disaster in the previous one :(
[17:38:26] <lynxlynxlynx> definitely
[17:38:36] <Avenger> heh?
[17:38:41] <Avenger> i thought that one is solved
[17:38:41] <lynxlynxlynx> my todo is on the wiki
[17:38:54] <lynxlynxlynx> it is, but it was solved post-release afair
[17:39:03] <Avenger> active in bg2, passive in bg1 :)
[17:39:06] <lynxlynxlynx> brad was working on the icon positioning
[17:39:14] <lynxlynxlynx> i asked fuzzie to fix leavearea stuff
[17:39:41] <lynxlynxlynx> all the rest is fuzzy
[17:39:51] <lynxlynxlynx> but if you need concrete goals, i can find them ;)
[17:40:16] <Avenger> can we disarm traps without stepping into them?
[17:40:41] <lynxlynxlynx> most of the time
[17:40:45] <lynxlynxlynx> i don't know if always
[17:40:52] <Avenger> yeah, i wonder
[17:40:55] <fuzzie> we shouldn't bother
[17:41:00] <fuzzie> we should just do what the original engine does
[17:41:13] <Avenger> or be better
[17:41:15] <fuzzie> although with a nice flag rather than checking the action id :P
[17:42:03] <Avenger> you mean, setting a flag in disarm and checking this flag in the pathfinding code?
[17:42:16] <fuzzie> well, checking the flag before activating the trap.
[17:42:32] <fuzzie> the original engine's region activation code checks if the current action on the actor is the disarm action, and if it is, then it ignores the actor.
[17:42:37] <Avenger> the original does that???
[17:42:47] <fuzzie> sure, it's pretty obvious
[17:43:11] <fuzzie> I don't have IDA with me but I'm pretty sure you named the functio.
[17:43:13] <fuzzie> function.
[17:43:20] <Avenger> O_o
[17:43:22] <Avenger> ok
[17:43:24] <Avenger> i will look at it
[17:43:30] <Avenger> i forget more than learn nowadays
[17:43:53] <fuzzie> the target point for the disasm action is the middle of the trap in the original engine, so it'd be really bad otherwise, as you can see by making a big trap :)
[17:44:07] <Avenger> yes
[17:44:10] <Avenger> that's the problem
[17:44:13] <fuzzie> but of course we don't have to be so silly
[17:45:32] <Avenger> well, i was thinking at checking the path, when the next step would be a traptodisarm->polygon->PointIn(...) then abort the pathfinding
[17:45:58] <Avenger> that needs an extra flag for the pathfinder and the trap's ID
[17:45:59] <fuzzie> yeah, but then the action has to poke deeply into the pathfinder
[17:46:10] <lynxlynxlynx> and it sounds expensive
[17:46:12] <Avenger> or the pathfinder deep into the polygon
[17:46:15] <fuzzie> so just having an IF_NOACTIVATETRAP seems the lazy way
[17:46:25] <Avenger> ohh
[17:46:29] <Avenger> well...
[17:46:37] <fuzzie> and I am lazy nowadays :-)
[17:47:36] <Avenger> but it shouldn't be just a bit flag
[17:47:49] <Avenger> if you walk into another trap other than the target, it should be fine
[17:48:01] <fuzzie> hmph, you're right
[17:48:07] <Avenger> so, you still have to have the flag and the trap id
[17:48:11] <fuzzie> you could have a TargetTrap field with the id
[17:48:13] <fuzzie> but that's kind of horrible
[17:48:26] <Avenger> not so horrible
[17:48:45] <Avenger> an IE creature is still a few times larger than ours :)
[17:48:50] <fuzzie> hehe
[17:48:57] <fuzzie> their size is not a goal! :)
[17:49:19] <Avenger> i know, but a bit and a dword won't bloat us either
[17:49:22] <fuzzie> of course even my android device has 1gb RAM nowadays, so..
[17:49:37] <fuzzie> if you have the field then I guess you don't need the flag
[17:49:51] <Avenger> hmm, ok just set it to invalid
[17:49:53] <fuzzie> since we never assign low globalids, so you can just set to 0 for no trap
[17:50:21] <Avenger> actually, that id could probably be needed for some trigger anyway
[17:50:42] <Avenger> i'm not sure we implemented all trap related triggers?
[17:52:16] <Avenger> well, i don't find much, except iwd2's InTrap shouldn't be set if the actor is right in the target trap we talk about
[17:53:38] <Avenger> btw, do we need to implement the said trigger?
[17:53:53] <lynxlynxlynx> lemme check if it is used
[17:54:02] <Avenger> i didn't find any use
[17:54:06] <Avenger> when i looked
[17:54:32] <Avenger> i don't know if it works at all
[17:54:50] <Avenger> i saw it is triggered in the code, but i have no idea how it works
[17:55:20] <lynxlynxlynx> i don't see any in scripts or dialogs
[17:56:26] <Avenger> it would be a worthy miniproject to find any remaining triggers/actions that we need to implement
[17:56:47] <Avenger> and actually reverse them :)
[18:02:23] <lynxlynxlynx> should be pretty easy if we assume modders have no imagination
[18:02:45] <lynxlynxlynx> we print all the non-stubbed actions and triggers on startup
[18:03:14] <lynxlynxlynx> the last one will likely be MultiplayerSync :)
[18:03:36] <Avenger> haha
[18:04:13] <lynxlynxlynx> saved trigger 85 (reset(o:object*)) doesn't exist, ignoring <-- the only other trigger in bg2
[18:04:39] <lynxlynxlynx> don't see any uses for this one either
[18:04:50] <Avenger> yeah, gotta see what is it
[18:05:24] <Avenger> i guess most of the stuff we don't know is stuff even bioware didn't know what to do with :D
[18:05:30] <lynxlynxlynx> oh, summoned is also there, but also unused
[18:05:38] <Avenger> i did summoned
[18:05:47] <Avenger> i think
[18:05:53] <lynxlynxlynx> [GameScript/WARNING]: Couldn't assign function to trigger: 151 summoned
[18:05:58] <Avenger> hmm
[18:06:19] <Avenger> it just needs a matcher function i guess
[18:06:54] <lynxlynxlynx> don't see any code
[18:07:19] <lynxlynxlynx> do we know when it should be triggered?
[18:08:30] <-- duckpunch has left IRC (Ping timeout: 272 seconds)
[18:09:39] <Avenger> LastSummoner is set, i guess :)
[18:09:54] <Avenger> but i'm unsure how it works
[18:10:35] <Avenger> i would return true if in the summoned creature and LastSummoner matches the object specification
[18:10:39] <Avenger> but that's just an idea
[18:11:07] <Avenger> could also need a trigger sent like with all non 0x4000 triggers
[18:11:55] <Avenger> i'm pretty sure it is just a few lines missing
[18:12:16] --> brada has joined #gemrb
[18:12:40] <brada> fuzzie: you can indeed use open gl in conjunction with SDL2 textures
[18:13:30] <brada> as for python 3 im sure apple will be the first to abandon python 2 so doubtless i will be the one dealing with that when it arises :p
[18:14:40] <brada> as for the next release I want to fix the status icons and chapter text both. I have them working locally, but the dependant font hacks i have need to be done right
[18:15:08] <lynxlynxlynx> ping me when you need result scrutiny
[18:15:15] <brada> of course
[18:15:19] <lynxlynxlynx> i still have a few tabs full of screenshots to compare with
[18:16:08] <brada> as for status icons im not sure what to do about centering them. i could make a blank icon for games that dont have one or we could shrink the rect so that centering is not necessary
[18:17:50] <lynxlynxlynx> oh right, they have to be centered to appear left aligned
[18:19:09] <lynxlynxlynx> using a smaller rect sounds much cleaner
[18:20:32] <brada> sort of. i already have the code that centers them beautiflly in bg2
[18:20:35] <lynxlynxlynx> some offsets are already available since we have to do them manually for the selection rect
[18:21:21] --> duckpunch has joined #gemrb
[18:21:42] <lynxlynxlynx> well, guiscripts can handle a lot, but it would be a shame to leave extra hacks in core
[18:22:37] <brada> i cant remember exactly, but i recall that normal text areas have no spacing between the lines while chapter screens do have line spacing
[18:28:45] <brada> lynx: i fear that different games would need their rects reduced by different amounts.
[18:30:04] <lynxlynxlynx> not much of a problem, it's defined in the scripts
[18:30:23] <brada> if so then its probably simpler to just craft a blank icon. the code for that is already done.
[18:35:27] <brada> also where is it defined in the scripts?
[18:35:40] <brada> the rect is created in Button.cpp
[18:38:04] <lynxlynxlynx> i meant the selection rect
[18:39:56] <lynxlynxlynx> look for Button.SetBorder in guicommonwindows
[18:41:00] <brada> so could we use the border rect for drawing the label? or would that break other buttons?
[18:42:15] <lynxlynxlynx> it's symetrical, but most of the buttons should have these offsets at 0
[18:42:58] <lynxlynxlynx> we use them in inventories and that's about it
[18:43:31] <lynxlynxlynx> maybe also for the equipped actionbar weapon, but i don't remember
[18:44:09] <lynxlynxlynx> if so, it could easily be done differently, since those buttons have a proper "second" or "third" frame
[18:44:41] <lynxlynxlynx> silly names, can never remember which is which
[19:14:19] <brada> lynx adding a single pixel is all that is required for BG1 & 2
[19:14:27] <brada> dont know about IWD series
[19:15:04] <lynxlynxlynx> cool
[19:15:18] <lynxlynxlynx> bg1 may still be off anyway
[19:15:25] <brada> why is that?
[19:15:31] <brada> i mean i just tested it
[19:15:36] <lynxlynxlynx> i mean the current selection rect
[19:15:58] <lynxlynxlynx> iwd2 on the other hand is the nice clean proper boy
[19:16:46] <brada> selection rect has no effect on how/where the status icons are drawn
[19:17:23] <lynxlynxlynx> i know
[19:18:26] <brada> if it is the same for all games we can just have a if (Picture && (Flags & IE_GUI_BUTTON_PORTRAIT) == IE_GUI_BUTTON_PORTRAIT) { // add 1 px to the x coordinate }
[19:20:13] <brada> or rather jsut (Flags & IE_GUI_BUTTON_PORTRAIT) == IE_GUI_BUTTON_PORTRAIT)
[19:21:46] <brada> kind of a hack, but so are all the other solutions
[19:32:07] <Avenger> lynx feel free to rename those second/third symbols. They are truly silly
[19:32:47] <-- Avenger has left IRC (Quit: bye!)
[19:33:28] <lynxlynxlynx> they're short though, it's hard to give a terse name to disabled-yet-clickable
[19:35:02] <brada> currently chapter text has a problem where text spontaneously appears due to text only being appended after the narrator is finished. in the original text is always appended even if narrator is still talking.
[19:35:25] <brada> we either need to do the same or push down appended text with blank lines
[19:35:44] <brada> i have code for the later, but it seems somewhat silly
[19:36:51] <brada> i was told the former was a bad idea for reasons i no longer recall
[19:37:50] <brada> let me know which solutions to pursue for both issues and i will see it done.
[19:37:51] <-- brada has left IRC (Quit: brada)
[19:38:32] <-- Earnestly has left IRC (Quit: WeeChat 0.3.9-dev)
[19:39:15] --> Earnestly has joined #gemrb
[19:40:19] <lynxlynxlynx> i don't understand, the narration only had some relevance to the scrolling speed
[20:02:15] --> edheldil_ has joined #gemrb
[20:03:07] --> brada has joined #gemrb
[20:03:20] <brada> lynx: let me try to clarify what i mean
[20:03:41] <brada> in the original the narrator will still be speaking the first paragraph when the second loads
[20:03:54] <brada> in gemrb we dont load the second paragraph untill narrator is done
[20:04:34] <lynxlynxlynx> so just increase the speed?
[20:05:02] <brada> afik there is no speed that will work for all paragraphs
[20:05:30] <brada> you get a speed for the first to second right, but second to third will be off
[20:05:35] <brada> pluss diffrent languages
[20:05:44] <brada> which means diffrent speeds
[20:05:54] <brada> so not really a good solution IMO
[20:06:32] <lynxlynxlynx> do you really thing they bothered with some wierd autosyncing in the original?
[20:07:07] <brada> i think in origial they load text as soon as the preceding paragraph is scrolled out
[20:07:09] <lynxlynxlynx> the whole chapter text is in one string
[20:07:14] <brada> is it?
[20:07:19] <brada> then why do we segment it?
[20:08:38] <lynxlynxlynx> we don't, it has embeded newlines
[20:08:45] <brada> no it is segmented
[20:09:00] <brada> textarea->appendText is called for each paragraph
[20:09:03] <lynxlynxlynx> well we add the paragraph markers if that's what you mean
[20:09:14] <lynxlynxlynx> oh, dunno about that
[20:09:36] <brada> and that is called only after the narration for the current paragraph is complete
[20:09:43] <brada> hence the problem
[20:10:29] <brada> so to match the original we should probably append all the text from the get go
[20:10:43] <brada> like i said tho, avenger indicated that would break something
[20:10:58] <lynxlynxlynx> the audio is not attached to the strings as usual btw
[20:11:07] <DrMcCoy> http://www.kickstarter.com/projects/obsidian/project-eternity :)
[20:11:39] <wjp> that's going nicely :-)
[20:11:39] <lynxlynxlynx> i can't think of anything, so best ask him directly
[20:12:05] <DrMcCoy> Linux at $2.2m
[20:12:16] <lynxlynxlynx> they should've put the mac goal further down
[20:12:24] <DrMcCoy> lol
[20:12:29] <lynxlynxlynx> plenty more zealots there
[20:12:52] <brada> i dont know about more zealots
[20:13:00] <brada> i know quite a large number of linux ones :p
[20:13:11] <DrMcCoy> >_> <_< Where? :P
[20:13:25] <brada> anyhow: line 235 of TextArea.cpp
[20:13:32] <lynxlynxlynx> if we append everything, how would autoscroll work?
[20:13:50] <brada> the diffrence between mac and linux zealots are mac ones are willing to spend money :p
[20:13:55] <lynxlynxlynx> i guess we'd just need to reset the position
[20:14:14] <DrMcCoy> brada: I think the Humble Bundles show otherwise
[20:14:30] <lynxlynxlynx> still, there are more ifans
[20:14:42] <lynxlynxlynx> so the net value would be higher
[20:15:00] <DrMcCoy> Yeah, they're still on my "to kill" list :P
[20:15:12] <DrMcCoy> Right after the Windowsians :P
[20:15:16] <brada> you sound like a zealot yourself when you say things like that
[20:15:27] <DrMcCoy> >_> <_< Who, me? Never!
[20:15:31] <brada> :p
[20:17:33] <DrMcCoy> Unfortunately, that kickstarter thing is pretty sparse on the details
[20:18:38] <DrMcCoy> Also, I hate how kickstarter still doesn't take PayPal
[20:33:55] <CIA-40> GemRB: 03avenger_teambg * r40ac27546f7e 10ielister/ielister.cpp: hardened item listing
[20:34:21] <-- Yoshimo has left IRC (Quit: Yoshimo)
[20:53:20] <-- brada has left IRC (Ping timeout: 245 seconds)
[21:11:42] --> brada has joined #gemrb
[21:29:27] <brada> i cant seem to find where the chapter audio is loaded...
[21:34:17] <-- CIA-40 has left IRC (Ping timeout: 240 seconds)
[21:42:23] <lynxlynxlynx> gemrb/GUIScripts/TextScreen.py ?
[21:43:16] <brada> doesnt seem to be
[21:43:50] <brada> at least not directly
[21:44:29] <lynxlynxlynx> the music is only the background?
[21:45:02] <brada> i thought.
[21:45:27] <brada> if i comment out the call to set FeedScroll and the out of text event then no sound plays tho
[21:45:47] <fuzzie> which audio?
[21:45:52] <fuzzie> the speech?
[21:45:57] <brada> yes
[21:46:13] <fuzzie> you'd think it'd be embedded in the text
[21:46:37] <fuzzie> as in, coupled to the resref
[21:47:04] <brada> must be
[21:47:48] <brada> so i dont know how to proceed here
[21:48:06] <fuzzie> well, you said it's all in one strref, and we split it up somewhere?
[21:48:32] <brada> well it appears to be split up to me
[21:48:53] <brada> in FeedScroll
[21:49:31] <fuzzie> yes, I don't see where
[21:49:57] <fuzzie> it just seems like it's being Appended strref-by-strref as the original has it
[21:50:02] <brada> all i really know is that only one paragraph is loaded at a time and it isnt untill the narrator finishes that the next paragraph is loaded
[21:50:15] <brada> i dont pretend to know how it functions underneeth
[21:50:25] <fuzzie> right, but the original must do it the same way
[21:51:09] <fuzzie> just you'd like to remove the IsSpeaking check in TextArea.cpp?
[21:51:17] <brada> no
[21:51:21] <brada> hang on
[21:51:41] <fuzzie> well, I mean, I realise that won't work
[21:52:44] <brada> i can tell you for a certainty that the original doent behave this way
[21:52:47] <brada> i jsut double checked
[21:52:53] <fuzzie> yeah, but you phrased it really oddly
[21:53:07] <brada> i tend to do that
[21:53:10] <fuzzie> oh I see, lynx phrased it really oddly :P
[21:53:18] <brada> he does as well i guess :p
[21:54:11] <fuzzie> see, now I'm just confused
[21:55:03] <brada> in the original the text for the next paragraph can be visible while the narrator is still speaking the current one
[21:55:41] <brada> gemrb waits till the narrator is done to load the next paragraph
[21:56:17] <lynxlynxlynx> the sound is not attached to the strref
[21:56:22] <brada> what complicates the situation is that loading the paragraph seems to be what starts the narration for the paragraph in gemrb
[21:56:29] <fuzzie> lynxlynxlynx: so how does it work?
[21:56:43] <lynxlynxlynx> i thought the music playlists we load manually
[21:56:47] <fuzzie> it isn't music
[21:57:02] <fuzzie> the '6' in the Append call in FeedScroll is IE_STR_SOUND|IE_STR_SPEECH
[21:57:39] <lynxlynxlynx> ok, then i guess someone has to check with dltcep
[21:57:43] <fuzzie> it would be good for those to be in GUIDefines.py
[21:57:44] <brada> thank you!
[21:57:53] <lynxlynxlynx> the string doesn't have the [sound] appended in the weidu dump
[21:57:58] <brada> wish we werent using magic numbers there
[21:58:22] <fuzzie> what is an example piece of chapter text?
[21:58:27] <fuzzie> brada: can you fix that? :P
[21:58:42] <brada> i will with the patch im working on
[21:59:35] <brada> i need to decouple text from speech tho...
[22:00:01] <fuzzie> that is not really trivial I guess.
[22:00:15] <lynxlynxlynx> 43624:@34089 = ~Chapter 1
[22:00:15] <lynxlynxlynx> 62070:@48007 = ~Chapter 1:~
[22:00:42] <lynxlynxlynx> the first one is the whole textarea worth, the second just the title
[22:00:50] <fuzzie> see @61359/61360/61361/61362
[22:01:02] <fuzzie> they're all one paragraph?
[22:01:12] <lynxlynxlynx> no
[22:01:23] <lynxlynxlynx> 91013:@73257 = ~Chapter 10
[22:01:27] <lynxlynxlynx> better example ^
[22:01:56] <fuzzie> yes, thanks
[22:02:18] <fuzzie> that's really quite odd
[22:03:05] <brada> im afraid i dont understand any of that...
[22:03:53] <fuzzie> oh wait
[22:03:57] <fuzzie> which game are you testing with? :P
[22:04:27] <brada> BG1
[22:05:24] <fuzzie> so, I guess bg1 is different
[22:05:34] <brada> how so?
[22:05:44] <fuzzie> because it *does* have the strings split up
[22:05:50] <brada> ah
[22:06:56] <fuzzie> so now I am less confused :-)
[22:07:15] <brada> well that makes one of us
[22:08:24] <brada> not sure how to remedy this situation
[22:08:30] <fuzzie> I guess you can trigger the speech from an appropriate point in the textarea, but you really have to do that change in the core.
[22:08:45] <-- wrotek__ has left IRC (Ping timeout: 246 seconds)
[22:09:05] <fuzzie> It seems weird that the original game would do that. Do you think it just plays all the speech in order with no gaps?
[22:09:49] <brada> i dont know how else it would work
[22:12:20] <fuzzie> and you really think the original doesn't do it as gemrb does?
[22:12:28] <brada> im certain of it
[22:12:51] <brada> i can clearly see text for another paragraph while the narrator is speaking the preceeding one
[22:12:57] <fuzzie> which chapter do you test with?
[22:13:05] <brada> the beginning
[22:13:11] <fuzzie> well, for me, that isn't the case
[22:13:14] <brada> so prologue i guess
[22:13:23] <fuzzie> the speech is just ahead of the text here
[22:13:25] <fuzzie> so that's annoying :P
[22:13:35] <brada> watch the entire thing
[22:13:42] <brada> not jsut the first paragraph
[22:14:11] <fuzzie> ah you're right, last paragraph is a separate strref again
[22:14:17] <fuzzie> hmph
[22:14:20] <brada> yeah
[22:14:23] <fuzzie> I guess we'll have to queue the audio then.
[22:14:29] <fuzzie> Doesn't look like there's any hacks in the original.
[22:15:18] <brada> i dont understand; why did they bother separating the paragraphs at all?
[22:15:36] <brada> the text i mean
[22:15:39] <fuzzie> well, they don't in bg2
[22:15:51] <fuzzie> but maybe in bg1 they were worried about the large audio buffers or something
[22:16:26] <fuzzie> it was designed to run on machines with 16mb RAM, so they were probably pretty squeezed
[22:16:58] <brada> what is our course of action then?
[22:17:13] <fuzzie> well, I am going to sleep
[22:17:17] <brada> ha ha
[22:17:24] <brada> ill ask again tomorrow then
[22:17:38] <fuzzie> but you can just keep a list of strrefs or something
[22:17:50] <fuzzie> the difficulty is just avoiding adding yet another hard-coded hack deep in the GUI code
[22:17:59] <fuzzie> but yes ask tomorrow please :) ninight
[22:18:04] <brada> good night
[22:42:25] <lynxlynxlynx> heh
[23:02:27] <-- lynxlynxlynx has left IRC (Remote host closed the connection)