#gemrb@irc.freenode.net logs for 21 Feb 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:13:09] <-- haad has left IRC (Quit: Leaving.)
[00:55:43] <-- edheldil_ has left IRC (Ping timeout: 244 seconds)
[02:34:23] <-- brad_a has left IRC (Quit: brad_a)
[03:02:07] --> brad_a has joined #gemrb
[05:03:34] --> alx3apps has joined #gemrb
[05:13:35] <-- brad_a has left IRC (Quit: brad_a)
[05:48:34] --> brad_a has joined #gemrb
[06:05:04] <-- brad_a has left IRC (Quit: brad_a)
[06:06:37] <tomprince> So, are we applying to GSoC.
[07:12:40] <fuzzie> good qu
[07:41:10] --> lynxlynxlynx has joined #gemrb
[07:41:10] --- ChanServ gives channel operator status to lynxlynxlynx
[08:02:09] <lynxlynxlynx> oh right, need to prepare everything
[08:02:26] <lynxlynxlynx> just having the discussion had great results though :)
[08:21:33] --> haad has joined #gemrb
[08:44:55] <-- kamui has left IRC (Read error: Connection reset by peer)
[08:49:28] --> edheldil_ has joined #gemrb
[08:54:41] <edheldil> lynxlynxlynx: what about adding unicode support to gsoc ideas? It would be extensive, but doable (and source of many many errors, to be sure)
[08:55:00] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[08:56:18] <lynxlynxlynx> sure, add whatever you think is doable
[09:05:48] --> haad1 has joined #gemrb
[09:09:15] <-- haad has left IRC (Ping timeout: 240 seconds)
[10:41:14] --> lostLinSoul has joined #gemrb
[10:42:02] <lostLinSoul> Hi all
[10:42:41] <lostLinSoul> Any way to adjust the positions of buttons on the gui?
[10:48:49] <fuzzie> how do you mean adjust?
[11:01:41] <lostLinSoul> Move the button to a new position
[11:02:23] <fuzzie> call SetPos on it
[11:04:27] <lostLinSoul> That will work from the python code?
[11:04:35] <fuzzie> sure?
[11:05:03] <lostLinSoul> Ta
[11:11:43] <-- edheldil has left IRC (Remote host closed the connection)
[11:20:59] <-- lostLinSoul has left IRC (Quit: Page closed)
[11:23:13] --> edheldil has joined #gemrb
[11:23:13] --- ChanServ gives channel operator status to edheldil
[11:53:35] <-- haad1 has left IRC (Quit: Leaving.)
[15:22:00] <CIA-28> GemRB: 03avenger_teambg * r23f59d775977 10gemrb/gemrb/core/ (4 files in 2 dirs): check if target is already affected by spell
[15:29:04] --> Avenger has joined #gemrb
[15:29:04] --- ChanServ gives channel operator status to Avenger
[15:29:59] <Avenger> tomprince: sdlvideo doesn't load for me anymore
[15:30:57] <fuzzie> what does it do?
[15:31:30] <Avenger> libversion is 0
[15:32:00] <fuzzie> eek?
[15:32:57] <fuzzie> hm, on windows?
[15:33:44] <fuzzie> because you need SDL12Video.cpp in your build now, of course
[15:35:03] <tomprince> 'tis what cmake is for
[15:46:34] <Avenger> hmm, what exactly i need for an sdlvideo plugin now? all .cpp in the dir, or just sdl12video, or sdl12video+sdlvideo
[15:46:46] <fuzzie> you need sdl12video+sdlvideo
[15:47:14] <Avenger> thanks
[15:48:27] <Avenger> meh now i forgot what i wanted in windows O_o
[15:51:25] <Avenger> shouldn't there be some name difference between the two sdl plugins?
[15:51:55] <fuzzie> well, you can't build them both
[15:52:00] <fuzzie> so not much point right now
[15:52:51] <Avenger> yeah, except that we see in logs that the guy is loading sdl 2.0 and claiming he has 1.2 and couldn't run the damn game :D
[15:53:29] <fuzzie> we do?
[15:53:38] <fuzzie> i mean, either you have one or the other :)
[15:53:52] <Avenger> isn't it in the log?
[15:53:58] <fuzzie> i don't see
[15:54:52] <Avenger> of course you don't see because it isn't in the lib name :P
[15:55:22] <Avenger> but i see this --> Loaded plugin "SDL Video Driver"
[15:55:37] <fuzzie> oh, you mean you could have some little message
[15:55:40] <fuzzie> but I think it's way too late then
[15:55:48] <lynxlynxlynx> are you sure that spell change is ok? it reads like it could make things more userfriendly
[15:56:03] <Avenger> what spell change
[15:56:04] <fuzzie> if someone managed to mess up the SDL versions then the library won't even load
[15:56:26] <tomprince> Well, once gentoo gets SDL2, i'll see about adding proper support for both.
[15:56:40] <lynxlynxlynx> this validtarget thing you just pushed
[15:57:27] <Avenger> there was no check earlier, at all. This will just make the validspelltarget (or whatever is the name of the trigger) at least ignore targets affected by the exact spell it is trying to cast again
[15:57:38] <Avenger> this is an iwd2 AI improvement
[15:57:42] <Avenger> nothing else uses it
[15:57:45] <lynxlynxlynx> can you still try to cast while being silenced or in a dead magic zone? Doesn't look like it
[15:57:57] <lynxlynxlynx> ah
[15:58:40] <Avenger> this shouldn't affect any gui, even in iwd2
[15:58:45] <Avenger> just the ai
[15:59:35] <Avenger> iwd2 ai has some complicated half buggy feature about picking the best offensive/defensive spell possible
[15:59:35] <lynxlynxlynx> MarkSpellAndObject("21122112210322112226222622222118",LastMarkedObject,7)
[15:59:39] <lynxlynxlynx> interesting params
[15:59:40] <Avenger> yeah, that's it
[16:00:00] <Avenger> it is a spell list, trying to find a spell that would actually has any effect on the target
[16:00:24] <Avenger> until now, the ai would pick always the same spell (first on list)
[16:00:35] <Avenger> now, i hope it would go through the list at least
[16:00:41] <lynxlynxlynx> :)
[16:01:13] <Avenger> the original engine had hardcoded rules about opcodes/actor states
[16:01:27] <Avenger> eventually i will implement that, somehow
[16:01:54] <lynxlynxlynx> what's missing for the spellbooks?
[16:02:28] <Avenger> well, i'm not sure the core knows about the class->spellbook index mapping yet
[16:02:31] <Avenger> maybe it does
[16:02:33] <lynxlynxlynx> in iwd2 you can have many, including that wierd divine granted ones
[16:03:00] <Avenger> the action button bar needs more love
[16:03:06] <Avenger> once it works, we can go deeper
[16:03:56] <lynxlynxlynx> yes :P
[16:04:17] <lynxlynxlynx> but the spellbooks are mostly work for the core and guima.py
[16:04:37] <Avenger> well, actually, the spellbook could be handled, at least you could list the spells in guispl
[16:05:02] <Avenger> i don't know what is needed in core to work
[16:05:21] <lynxlynxlynx> do we handle all the extra types already?
[16:05:42] <lynxlynxlynx> i remember seeing some code for the spell types, but books should be the same, i guess
[16:06:27] <Avenger> at least ieIWD2SpellType lists the book names
[16:06:54] <Avenger> now the trick is to access the correct book based on the class
[16:07:12] <Avenger> that is something for the guiscript, i think
[17:01:47] --> pmjdebru1jn has joined #gemrb
[17:02:30] <-- Avenger has left IRC (Quit: bye!)
[17:02:34] <-- pmjdebruijn has left IRC (Ping timeout: 245 seconds)
[17:08:31] --> kettuz has joined #gemrb
[17:39:53] --> Yoshimo has joined #gemrb
[17:50:19] --> edheldil_ has joined #gemrb
[17:55:14] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[17:59:52] --> brad_a has joined #gemrb
[18:27:02] --> SiENcE has joined #gemrb
[18:34:20] <brad_a> alx3apps: are you able to build against SDL 2?
[18:39:10] <alx3apps> brad_a: I doubt that I'll be able do it quickly. I still haven't look at audio part you wrote. And SDL 2 is not in pelya's project (so I need to add it, see it won't compile etc. - as it were with sdl1.3). I'm going to return to this on weekend.
[18:40:06] <brad_a> well SDL 1.3 is now SDL 2.0 :-p
[18:40:27] <brad_a> but im planning on rewriting touch input to be android compatible
[18:40:56] <brad_a> but if there isnt going to be a release of GemRB on android with SDL 2 then im in no hurry
[18:48:26] --> kingron has joined #gemrb
[18:49:11] <alx3apps> brad_a: so you are in no hurry :). Maybe Beholder will appear and fix it, as usual. I'm interested in working on gemrb, but I'm not even near android native development now. And not going to rush. For now i'm looking at proper buildslave config (wrote about it yesterday). And after years of GC languages C programming is really fun (valgrind he-he). The next milestone is to learn about proper libs linking.
[18:57:07] <brad_a> i havent even seen beholder around. but even if we can build android against SDL 2 it may be intollerably slow
[18:57:20] <brad_a> so we may want 2 android builds or something
[18:59:28] <alx3apps> brad_a: I did sdl-1.3 builds (from old pelya's version) for some time and haven't seen speed problems (testing only by one-minute load-click-close though). Does SDL2 differ from SDL1.3 so much?
[18:59:48] <brad_a> no
[18:59:56] <brad_a> it differs from 1.2 a lot tho
[18:59:57] <fuzzie> they just removed the backwards-compatibile APIs
[19:43:51] <-- kettuz has left IRC (Quit: Leaving)
[19:58:14] --> edheldil_ has joined #gemrb
[21:34:59] <-- Yoshimo has left IRC (Quit: Yoshimo)
[21:41:47] <-- alx3apps has left IRC (Quit: Leaving.)
[21:48:42] --> gfag has joined #gemrb
[21:49:17] <gfag> evening ladies and gents
[21:51:36] <lynxlynxlynx> oj
[21:54:18] <gfag> Anybody here got any ideas how to get BG2 SOA running under android 2.3?
[21:58:05] <fuzzie> usually there isn't anyone here who knows about android, forum is a much better bet for help
[21:59:04] <fuzzie> hm, buy-one-get-one-free sale on gog for D&D games, with ToEE thrown in with any purchase too
[22:01:50] <gfag> ouch! you guys were my best bet... I've already read every single thread available... wasted hours of my life! :D
[22:02:46] <fuzzie> well: if you ask a more specific question then maybe someone will wake up and help
[22:11:31] <gfag> nah! i'll be better bothering the forum... but thanks anyway.
[22:13:22] <gfag> goddamn i was looking to play this awesome game everywhere where i want!
[22:21:30] <edheldil_> just ask, don't ask to ask
[22:21:42] <edheldil_> Our event handling is a mess
[22:22:12] <fuzzie> which bit?
[22:24:00] <edheldil_> gamescreen specifically, but I think it's not confined there :)
[22:24:26] <lynxlynxlynx> the input part is in flux at the moment
[22:24:32] <edheldil_> fuzzie: bad luck, I have already bought all of them, I think
[22:24:33] <fuzzie> well, the key stuff in gamecontrol is a mess
[22:25:17] <lynxlynxlynx> yeah, we only support shortcuts in gamecontrol, not anywhere else (if i don't count default/cancel)
[22:25:19] <fuzzie> but the mouse stuff doesn't look so bad
[22:27:14] <fuzzie> although 'git blame' blames me for the mouse stuff and not the key stuff so I may be biased actually :-p
[22:47:04] <edheldil_> well, speaking of mouse in gamecontrol - would you care to comment on https://github.com/edheldil/gemrb/commit/5ac53f147430948b8d0e37b09a06422b81724f96 ?
[22:48:24] <tomprince> evil evil use of the variable dictionary.
[22:48:28] <brad_a> dos formation rotation still work with those changes?
[22:48:40] <brad_a> oh nevermind i see you addressed that
[22:49:23] <brad_a> tomprince: wasnt it doing that before?
[22:49:42] <tomprince> That doesn't change the fact that it is evil evil evil.
[22:49:51] <brad_a> he he yeah
[22:50:06] <brad_a> has anybody ever heard of this problem: http://forums.gibberlings3.net/index.php?showtopic=24046
[22:50:17] <brad_a> i dont have BG1 installed to check
[22:50:59] <-- basse has left IRC (Ping timeout: 245 seconds)
[22:51:27] <brad_a> tomprince: i think we need a way to do a version sanity check for plugins
[22:51:56] <brad_a> unless it would be hard
[22:54:19] <-- kingron has left IRC (Remote host closed the connection)
[22:54:42] <edheldil_> brad_a: I have bg1, but not a suitable save
[22:55:10] <edheldil_> tomprince: is there a better way now than var dict?
[22:56:22] <tomprince> That is a good question. It seems you just want to pass a pair to the function here, right?
[22:56:46] <edheldil_> yes ...
[22:58:14] --> basse has joined #gemrb
[23:01:29] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:15:14] <edheldil_> brad_a: I am not sure you are responsible, but is there a reason not to include all sprites when loading a BAM font?
[23:15:36] <brad_a> im sure i am responsible
[23:15:48] <brad_a> tho can you be more specific?
[23:16:01] <brad_a> are you talking about the ability to limit them in fonts.2da?
[23:16:05] <fuzzie> wasn't the idea to restrict the loaded glyphs for sprites which might be replaced by ttf?
[23:17:18] <brad_a> the primary reason for limiting glyphs is because if you are using a TTF font you only need/want a small subset of the entire font
[23:17:26] <brad_a> and BAM fonts jsut came along for the ride
[23:17:37] <edheldil_> as far as I can see, the bam fonts limits in ps:t are set to 33:255
[23:17:51] <edheldil_> does it include space?
[23:18:06] <brad_a> i ont remember :)
[23:18:32] <fuzzie> does pst need space?
[23:18:50] <edheldil_> sure :)
[23:18:55] <fuzzie> what does it use it for?
[23:19:08] <edheldil_> for space between words :)
[23:19:19] <brad_a> well obviously something is doing that :p
[23:19:20] <fuzzie> sure, but why do you need the sprite?
[23:19:23] <fuzzie> bad sizing?
[23:19:27] <edheldil_> yes
[23:19:36] <edheldil_> we have a too narrow space
[23:19:39] <brad_a> for just PST or everything?
[23:19:45] <fuzzie> doing the calculation like that is indeed a bit silly
[23:19:57] <fuzzie> we should pull it from the sprite in BAM mode
[23:20:05] <edheldil_> exactly
[23:20:10] <brad_a> the calculation should only be done if space isnt included in the range
[23:20:23] <brad_a> so how about just change the starting char to 32?
[23:20:30] <brad_a> and everybody is hap i thinnk
[23:20:33] <brad_a> happy
[23:20:49] <fuzzie> well, the hard-coded calculation is probably not good in general
[23:20:54] <fuzzie> but I guess it's kind of fiddly
[23:21:31] <fuzzie> hum.
[23:21:47] <edheldil_> yeah, it worked :)
[23:21:54] <brad_a> there we go :)
[23:22:06] <brad_a> so just change all the font.2da files to start at 32 i guess
[23:22:13] <edheldil_> the line spacing is bad as well, I think :)
[23:22:31] <brad_a> what controls that?
[23:22:47] <brad_a> probably maxheight?
[23:23:29] <brad_a> i know bams had some special glyph at the beginning we were using some places but that didnt work with ttf
[23:23:36] <brad_a> so i changed it to maxheight
[23:23:46] <brad_a> im sure we can work something out
[23:25:53] <edheldil_> I think you are right on both counts
[23:26:16] <-- gfag has left IRC (Ping timeout: 245 seconds)
[23:26:21] <edheldil_> the text rendering is full of nice hardcoded magic numbers :)
[23:26:41] <brad_a> well not so many as there used to be ;)
[23:26:42] <tomprince> edheldil_: http://github.com/tomprince/gemrb/commits/point-to-python <--- untested, but it compiles
[23:29:14] <edheldil_> tomprince: is not it a bit ad hoc? Avenger will want to revert it :)
[23:29:41] <gembot> build #566[point-to-python] of cmake clang++ is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/cmake%20clang%2B%2B/builds/566 blamelist: Avenger, tomprince, edheldil
[23:30:20] <edheldil_> brad_a: set line spacing in fonts.2da for ttf fonts
[23:30:21] <tomprince> edheldil_: In what way is that adhoc?
[23:30:59] <gembot> build #164[point-to-python] of osx-xcode-binary is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/osx-xcode-binary/builds/164 blamelist: Avenger, tomprince, edheldil
[23:31:30] <edheldil_> it's more or less specific for this one case. But if you think it's better than using vars, I will use it
[23:31:55] <tomprince> After I fix the compile errors from rebasing. :)
[23:32:21] <gembot> build #185[point-to-python] of osx-test is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/osx-test/builds/185 blamelist: Avenger, tomprince, edheldil
[23:32:27] <tomprince> gembot: mute
[23:32:27] <gembot> Shutting up for now.
[23:32:52] <brad_a> ed: yes ill do that
[23:34:30] <CIA-28> GemRB: 03tom.prince * r9dfa5452554f 10gemrb/gemrb/plugins/GUIScript/ (GUIScript.cpp GUIScript.h):
[23:34:30] <CIA-28> GemRB: GUIScript: Factor out common code from RunFunction and CallbackFunction.
[23:34:30] <CIA-28> GemRB: They (almost) duplicated the same logic for finding and calling a
[23:34:30] <CIA-28> GemRB: function (except RunFunction could look in either the current or
[23:34:30] <CIA-28> GemRB: other modules). Rename CallbackFunction to RunFunction, while
[23:34:31] <CIA-28> GemRB: extending it to support other modules, and call this from the
[23:34:32] <CIA-28> GemRB: original RunFunction.
[23:34:32] <CIA-28> GemRB: 03tom.prince * r863aeda22cbf 10gemrb/gemrb/ (3 files in 2 dirs): GUIScript: Add a function call python with a Point.
[23:35:37] <edheldil_> btw, I was thinking about unicode support - does it make sense to support unicode with BAM fonts?
[23:35:48] <brad_a> sure
[23:37:02] <brad_a> tho i dont know exactly what you mean
[23:37:20] <edheldil_> are you already able to print unicode strings with TTF fonts in gemrb?
[23:37:26] <edheldil_> tomprince: thanks
[23:37:38] <tomprince> edheldil_: I find stuffing all comuncation between C++ and python in a big dict more adhoc.
[23:38:04] <tomprince> Certainly, this is only currently used in one place, but I suspect it won't be the last.
[23:38:59] <brad_a> ed: it doesnt matter if a you are using bam or ttf font. all that is happening is we take a given character as an index to an array to pull out a sprite
[23:39:30] <edheldil_> ok, I thought that you might have more logic there
[23:39:55] <-- SiENcE has left IRC (Quit: cya)
[23:41:04] <brad_a> i think we need to use different string functions to support multibyte cahracters
[23:41:13] <brad_a> thats problem 1
[23:41:37] <brad_a> then we would need a means to dynamically convert dialog strings and such to unicode
[23:42:20] <brad_a> TTF plugin should itself be unicoe ready
[23:43:17] <tomprince> Looking at the GameControl code, I get the feeling that more of it should be in python.
[23:43:36] <brad_a> ideally we can convert to unicode without adding a dependency
[23:44:16] <edheldil_> so my idea for Font class is: vector of glyphs ; ucs2 to glyph index table ; legacy 8 bit encoding -> unicode translation table (possibly merged with the last one on SetEncoding()
[23:44:40] <brad_a> why a vector?
[23:44:48] <edheldil_> let's ignore compund characters
[23:45:21] <tomprince> Are there people looking for unicode support?
[23:45:26] <brad_a> sort of
[23:45:28] <edheldil_> so that you can build it per partes
[23:45:45] <brad_a> what?
[23:45:48] <edheldil_> that's why I was asking
[23:46:35] <tomprince> Well, I'm just thinking that it would be useful to talk to the people who are going to be using it before designing it, perhaps.
[23:47:03] <edheldil_> well, I was thinking about building a font in several steps from several bams, but that does not really matter
[23:47:13] <brad_a> i have seen 1 person on the forums since TTF came out wanting to use it with some non ASCII language
[23:47:52] <brad_a> i agree that maybe we can ignore BAMs at least for the time being
[23:47:59] <tomprince> How do they expect it to behave?