#gemrb@irc.freenode.net logs for 9 Mar 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:17:51] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[00:20:14] <-- edheldil_ has left IRC (Ping timeout: 246 seconds)
[02:42:03] <-- |Cable| has left IRC (Ping timeout: 248 seconds)
[02:42:54] --> |Cable| has joined #GemRb
[05:48:19] <-- jschall has left IRC (Remote host closed the connection)
[06:01:28] --> raevol has joined #GemRb
[06:04:24] <-- tomprince has left IRC (Ping timeout: 255 seconds)
[06:26:23] --> tomprince has joined #GemRb
[06:40:34] --> Aki_ has joined #GemRb
[06:45:40] <-- raevol has left IRC (Quit: Leaving.)
[06:50:20] --> raevol has joined #GemRb
[07:14:16] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[07:35:54] <Aki_> Hello world ?
[07:38:00] <pupnik> hallo Aki_
[07:38:06] * ar rewrites Aki_ in ruby
[07:38:22] <ar> you're now: puts "Hello world ?"
[07:39:23] <Aki_> :)
[07:39:58] <Aki_> So you guys are also early morning guy ? Or is it still late at night ?
[07:43:05] <pupnik> time has no meaning. to a true student, a year is as a day.
[07:43:58] <Aki_> Wish I could still be student
[07:44:39] <Aki_> Never attended lessons, can code as much as I wanted
[07:45:33] <Aki_> Are you student pupnik ?
[07:48:49] --> Drakkar has joined #GemRb
[07:48:57] <Aki_> Hi drakkar
[07:50:58] <ar> i've been a student, but got a job as a sysadmin
[07:53:33] <Aki_> Same for me
[07:53:37] <Aki_> + familly
[07:55:21] <-- raevol has left IRC (Quit: Leaving.)
[07:56:08] <ar> your whole family got sysadmin jobs? ;)
[07:56:40] <fuzzie> a terrible fate :)
[07:57:05] <Aki_> We're all RHSA ;) Including 18mth son
[07:57:06] <Aki_> :p
[07:58:48] --> Aki- has joined #GemRb
[07:59:26] <-- Aki_ has left #GemRb
[08:02:40] --> Bo_Thomsen has joined #GemRb
[08:22:20] --> adominguez has joined #GemRb
[08:27:34] --> lubos has joined #GemRb
[08:45:45] --> edheldil_ has joined #GemRb
[08:48:35] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[08:49:13] --> SiENcE has joined #GemRb
[08:50:03] <-- edheldil_ has left IRC (Ping timeout: 248 seconds)
[09:46:04] <-- |Cable| has left IRC (Remote host closed the connection)
[10:00:49] <edheldil> Morning, all!
[10:02:02] <Aki-> hi
[10:02:17] <wjp> morning
[10:02:53] <edheldil> How, is your savegame editor progressing, Aki?
[10:04:53] <Aki-> Got a dayjob ;) Coding once back home
[10:05:14] <Aki-> But already stratind dissecting saves in perl
[10:05:30] <Aki-> unsure in what I will write the GUI
[10:05:33] <Aki-> Java probly
[10:06:56] <edheldil> had you written the dissector in Python, I could add it to iesh ;-)
[10:07:14] <Aki-> What is iesh ?
[10:07:19] <edheldil> btw, can't NI open gam files? Just asking ..
[10:07:35] <Aki-> Sure it can
[10:07:43] <Aki-> Have you eve tried shadowkeeper ?
[10:07:48] <edheldil> nope
[10:07:52] <Aki-> The good old windows app
[10:08:12] <Aki-> http://www.gram.pl/upl/artykul/20081107203818.jpg
[10:08:22] <Aki-> lil' more friendly than NI
[10:09:04] <edheldil> neat
[10:09:32] <edheldil> iesh is ie shell, my horribly incomplete python lib over IE files
[10:09:51] <Aki-> got a url ? git/svn repo ?
[10:10:32] <edheldil> it's in gemrb's git, just iesh.git at the end, I think
[10:10:58] <Aki-> will try to have a look
[10:11:25] <edheldil> http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/ie_shell;a=summary
[10:12:01] <Aki-> Are you jara ou jeroslav ? :D
[10:12:04] <edheldil> no worries, it's more and more stale each day :)
[10:12:27] <edheldil> actually Jaroslav
[10:12:46] <Aki-> Sorry for typo
[10:12:55] <edheldil> np
[10:15:58] <Aki-> Got to go, see you later (Brand new IRC BNC, don't know how it'll behave)
[11:20:35] <Aki-> Back
[11:52:05] <-- SiENcE has left IRC (Quit: @all: cya)
[12:00:31] --> hanicka has joined #GemRb
[12:01:00] <hanicka> hi, i have question about static compiling of gemrb on apple
[12:01:33] <fuzzie> it's disabled due to symbol issues, right?
[12:02:22] <hanicka> i managed to compile without errors
[12:03:02] <hanicka> but i don't understand why pluginmgr trying load plugins when there is'nt any
[12:03:54] <fuzzie> that shouldn't be a problem
[12:04:18] <fuzzie> the idea is that you can have a mixed build, some static plugins and some dynamic plugins
[12:07:08] <hanicka> make install do'nt create any plugin directory, gemrb fail on: [PluginMgr]: Loading Plugins from /usr/local/plugins/ [Core]: Plugin Loading Failed, check path...[ERROR]
[12:07:09] <fuzzie> but i imagine that if you sabotage our cmake setup to do builds on OS X, you end up with no static plugins in your binary, because none of the plugin symbols are used.
[12:08:14] <fuzzie> it's not that it's missing the plugin *files*, it's that you have no plugins at all, not even built-in.
[12:09:15] <fuzzie> you might try '-all_load' as a linker flag.
[12:10:32] <fuzzie> (in the TARGET_LINK_LIBRARIES bit you presumably already modified in gemrb/CMakeLists.txt)
[12:11:55] <hanicka> ok, i will try...
[12:12:51] <fuzzie> i don't know how to pass linker stuff on OS X, but i am guessing -Wl, like the GNU toolchain option we already use
[12:16:14] <fuzzie> of course this assumes that you did already modify the file and link the plugins :)
[12:30:28] <-- hanicka has left IRC (Remote host closed the connection)
[12:31:23] --> hanicka has joined #GemRb
[12:46:08] <hanicka> ok, i get working static build on mac osx :)
[12:46:30] <hanicka> few warnings, but ok :)
[12:49:53] --> Demitar has joined #GemRb
[12:52:13] <hanicka> ohh, sound in movies!! nice (openal is'nt working, but sdlaudio yes)
[12:52:21] <-- devurandom has left IRC (Ping timeout: 260 seconds)
[12:55:30] --> devurandom has joined #GemRb
[13:18:11] --- Aki- is now known as Akileos
[13:19:25] <Akileos> hanicka: openal not working ?
[13:20:10] <fuzzie> openal on OS X is a bit of a nightmare.
[13:20:12] <Akileos> I'm using it without any problem
[13:20:27] <Akileos> That's weird
[13:20:47] <Akileos> if using SDLAudio no sound at all
[13:21:02] <Akileos> is my computer cursed ? :D
[13:21:13] <fuzzie> SDLAudio is really inferior, it doesn't do ambient sounds
[13:21:35] <fuzzie> i wrote it just because there's no nice embedded version of openal
[13:21:47] <hanicka> yep, SDLAudio don't have ambient sound, but music and video! :)
[13:22:05] <Akileos> Well OpenAL works fine here
[13:22:25] <fuzzie> but you are on some desktop win/linux platform i assume :P
[13:22:41] <Akileos> nope, OSX too
[13:23:01] <fuzzie> i think there's an openal library in macports which is fine
[13:23:09] <Akileos> That's the one I used
[13:23:17] <fuzzie> but then you have gemrb linked against macports, which is not so helpful
[13:23:31] <Akileos> why would it be linked to macports ?
[13:24:05] <hanicka> system openal have ambient, UI sound, bud not music and videos
[13:24:18] <Akileos> I do have music + vid
[13:24:24] <fuzzie> oh, hm, maybe not, macports openal looks like it might build nice framework stuff nowadays
[13:24:48] <hanicka> footstep sound is'nt working on both
[13:25:00] <hanicka> i will try macports openal
[13:25:08] <fuzzie> if it works for Akileos, it sounds like a good idea to try
[13:26:00] <fuzzie> useful to know these things
[13:26:21] <fuzzie> if you could submit patch for static_link people might appreciate it, also
[13:28:03] <hanicka> fuzzie: yes, just few warnings to fix
[13:28:48] <hanicka> after i will try sdl1.3, if it work I can port to iOS
[13:29:11] <fuzzie> it works with SDL 1.3 but you have to update a couple of the function calls, i think palette ones
[13:30:26] <fuzzie> there are probably warnings about alignment in bikplayer still
[13:35:56] <Akileos> I crahed everything :D Can't run screenuim + gemrb
[13:46:23] <Akileos> Well looks like I also have some fooststeps
[13:46:37] <Akileos> Will upload vid to youtube
[13:47:05] <hanicka> How you force cmake to get openal from macports instead of preferring system one
[13:47:07] <hanicka> ?
[13:48:34] <wjp> by setting OPENALDIR, if I read this right
[14:01:30] <Akileos> hanicka: Are you building from GIT ?
[14:01:39] <hanicka> yes
[14:07:02] <hanicka> i have few warning like this:
[14:07:02] <hanicka> ld: warning: Palette::Release() has different visibility (hidden) in plugins/BAMImporter/BAMImporter.a(BAMImporter.o) and (default) in core/libgemrb_core.a(AnimationFactory.o)
[14:08:17] <wjp> hm, an inline
[14:09:08] <hanicka> and one: /usr/bin/ranlib: file: libgemrb_core.a(snprintf.o) has no symbols
[14:10:01] <wjp> yeah, snprintf.o is empty on many platforms
[14:13:02] <hanicka> where is mouse scrolling? i think about add scrolling left/right (on mac mouse/touchpad)
[14:13:21] <Akileos> http://www.youtube.com/watch?v=Uvs7yQjgbU8
[14:14:25] <hanicka> Akileos: nice :)
[14:14:32] <Akileos> So you experts can you tell me what's missing :p
[14:14:41] <Akileos> Think I have voice, footspteps
[14:14:45] <Akileos> and videos
[14:14:55] <hanicka> music?
[14:15:24] <Akileos> Didn't notive it was missing
[14:15:28] <Akileos> maybe not
[14:15:35] <Akileos> ambient sound
[14:16:00] <hanicka> music in main menu definitely
[14:16:05] <Akileos> maybe no music, you may be right
[14:17:31] <Akileos> So what should I do ?
[14:17:32] <Akileos> :D
[14:20:07] <wjp> take a look at audio volumes and any messages in the terminal
[14:24:17] <Akileos> [ResourceManager]: Searching for songlist.2da...[Override]
[14:24:18] <Akileos> [MUSImporter]: Loading /Volumes/BG/BG2/music/THEME.MUS...[FOUND]
[14:24:18] <Akileos> [ResourceManager]: Searching for THEME/THEMEA... THEME/THEMEA.acm...[Music]
[14:24:18] <Akileos> [OpenAL]: Unable to create music source: 40963 [ERROR]
[14:24:18] <Akileos> Playing: THEME/THEMEA
[14:25:04] <hanicka> I have same error with my openal...
[14:25:27] <Akileos> but u have the music ?
[14:26:33] <wjp> hm, 40963 is invalid value
[14:26:40] <hanicka> music yes, video audio no, ui sound
[14:26:46] <hanicka> yes
[14:26:48] <hanicka> :)
[14:27:23] <wjp> ...which is a rather weird error code in that location
[14:29:14] <wjp> does it say anything about "Allocated ... streams" earlier in the logs?
[14:29:15] <hanicka> patch fot static compile on mac (+ small fix of sdlaudio compile, on mac sdlaudio headers is in different ".framework" path)
[14:29:16] <hanicka> http://nimue.cz/gemrb-static.txt
[14:29:48] <wjp> hanicka: thanks; could you upload it to our patch tracker on SF so we don't forget?
[14:29:56] <hanicka> wjp: ok
[14:33:09] <Akileos> http://pastebin.com/rFBFuKNA
[14:34:03] <wjp> aha, there's an earlier error already
[14:34:11] <wjp> Unable to create music buffers: 40963
[14:35:51] <hanicka> ok, patch uploaded :)
[14:36:16] <wjp> thanks
[15:38:04] <MikeChelen> anyone know how to fix this issue with trolls? https://sourceforge.net/tracker/?func=detail&aid=3200851&group_id=10122&atid=110122
[15:41:15] <ar> MikeChelen: btw, doesn't it lead to infinite experience points?
[15:56:41] <edheldil> I fondly remember 'infinite white cube bug' in Lands of Lore :-)
[15:57:18] <ar> i remember trolls being quite hard to kill in nethack/slashem, but not invincible
[16:03:42] <fuzzie> you have to provide ctrl-m output on a troll
[16:03:57] <fuzzie> (and you don't get infinite xp because they aren't dying)
[16:05:14] <Akileos> Work time is over ! Let's have some savegame fum
[16:06:05] --> pupnik_ has joined #GemRb
[16:07:47] <-- pupnik has left IRC (Read error: Operation timed out)
[16:08:23] <Akileos> That fun, it's not actually the first storm that crashes game but the second
[16:08:28] <Akileos> In saradush
[16:08:38] <Akileos> So can't even kill the npc to avoid it
[16:26:35] --> lynxlynxlynx has joined #GemRb
[16:26:35] <-- lynxlynxlynx has left IRC (Changing host)
[16:26:36] --> lynxlynxlynx has joined #GemRb
[16:26:36] --- ChanServ gives channel operator status to lynxlynxlynx
[16:31:15] <MikeChelen> ar: haha not sure, it doesn't seem to ever register as dead
[16:32:47] <MikeChelen> fuzzie: how do you enable cheat keys to let ctrl-m work?
[16:33:06] <fuzzie> it's in gemrb.cfg
[16:33:48] <MikeChelen> ah found it
[16:35:52] --> Maighstir has joined #GemRb
[16:40:49] <-- lubos has left IRC (Quit: Leaving.)
[16:50:42] <-- hanicka has left IRC (Remote host closed the connection)
[17:12:57] --> Bo_Thomsen has joined #GemRb
[17:48:36] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[17:52:04] <-- Demitar has left IRC (Quit: Burn the land and boil the sea, you can't take the sky from me!)
[18:02:23] <-- adominguez has left IRC (Quit: Saliendo)
[18:14:50] --> |Cable| has joined #GemRb
[18:43:17] --> Demitar has joined #GemRb
[18:46:53] <-- |Cable| has left IRC (Read error: Connection reset by peer)
[18:48:18] --> |Cable| has joined #GemRb
[19:01:38] <tomprince> hanicka: does it work without the -fvisibility?
[19:03:34] <tomprince> who isn't around.
[19:04:11] <fuzzie> thinking of the warnings?
[19:26:56] <-- |Cable| has left IRC (Remote host closed the connection)
[19:28:17] --> |Cable| has joined #GemRb
[19:28:38] <-- |Cable| has left IRC (Client Quit)
[19:57:06] --> edheldil_ has joined #GemRb
[19:59:27] --> Avenger has joined #GemRb
[19:59:28] --- ChanServ gives channel operator status to Avenger
[19:59:48] <Akileos> Hi Avenger
[20:00:00] <Avenger> fuzzie: i think the problem with trolls is caused by ChangeAnimationNoEffect. It heals the troll fully
[20:00:03] <Akileos> I've been told you're android guru in there
[20:00:05] <Avenger> hello akileos
[20:00:11] <Avenger> no, that's not me
[20:00:29] <Akileos> Too bad :)
[20:00:33] <Avenger> i probably know less about android than you
[20:01:03] <Akileos> So you maybe know why I can pass Saradush :D
[20:01:17] <Akileos> can't
[20:01:44] <Avenger> difficult battle?
[20:02:05] <Akileos> Crashing everytime
[20:02:17] <Akileos> the red storm makes gemrb crash
[20:02:18] <Avenger> ouch
[20:02:43] <Akileos> null pointer on self, if I understood what they told me yesterday
[20:02:48] <Avenger> which gemrb version
[20:03:06] <Avenger> anyway, first i try to find the troll problem
[20:03:07] <Akileos> git latest
[20:03:30] <Akileos> And giveStartEquipement in tob doesn't work either
[20:04:17] <fuzzie> someone should just revert the commit for that
[20:04:42] <fuzzie> 62dd9fee, specifically
[20:05:33] <fuzzie> just some 'git revert 62dd9fee' should do it, but i can't test it right now
[20:09:58] <Avenger> yeah, i found the code where it copies the hp from the old actor to the new one
[20:10:07] <fuzzie> how helpful :P
[20:10:20] <Avenger> i mean, i see it in the original
[20:10:22] <Akileos> [GUIScript]: Runtime Error:
[20:10:22] <Akileos> Traceback (most recent call last):
[20:10:22] <Akileos> File "/Users/Aki/Devel/gemrb/build/gemrb/GUIScripts/bg2/CharGen7.py", line 23, in OnLoad
[20:10:22] <Akileos> CharGenCommon.DisplayOverview (7)
[20:10:23] <Akileos> File "/Users/Aki/Devel/gemrb/build/gemrb/GUIScripts/bg2/CharGenCommon.py", line 147, in DisplayOverview
[20:10:23] <Akileos> if GemRB.GetPlayerStat (MyChar, IE_SEX) == 1:
[20:10:23] <Akileos> RuntimeError: Cannot find actor!
[20:10:41] <fuzzie> ok that sounds not so good
[20:10:49] <Avenger> bug in chargen?
[20:11:23] <Akileos> reverted to 62dd9fee
[20:12:06] <fuzzie> you did 'git revert', or you moved back to 62dd9fee?
[20:12:37] <Akileos> just git revert
[20:12:48] <Akileos> well ok, not "to"
[20:12:59] <fuzzie> sorry, just wanted to be sure
[20:13:17] <fuzzie> because the commit only changes CharGenEnd, so i think can't cause that bug
[20:13:41] <Avenger> anyone can commit? if yes, just copy the hp in changeanimation from the old actor to the new actor :) that fixes the troll problem
[20:13:54] <fuzzie> hm
[20:14:09] <fuzzie> i can blindly commit
[20:15:21] <Avenger> i tell you where
[20:15:23] <fuzzie> you want the fix in ChangeAnimationCore?
[20:15:25] <fuzzie> thanks
[20:15:36] <Avenger> tar->SetOrientation(src->GetOrientation(), false );
[20:15:38] <Avenger> tar->BaseStats[IE_HITPOINTS]=src->BaseStats[IE_HITPOINTS];
[20:15:40] <Avenger> src->DestroySelf();
[20:15:41] <Avenger> yes
[20:15:45] --> kettuz has joined #GemRb
[20:15:51] <Avenger> i think this will fix it
[20:16:43] <Avenger> i will test it now :)
[20:18:54] <fuzzie> Akileos: can you reproduce that error?
[20:22:08] <fuzzie> Avenger: did it work? :P
[20:22:25] <Avenger> testing
[20:22:48] --> Bo_Thomsen has joined #GemRb
[20:23:14] <Avenger> fuzzie: there is another bug i think
[20:23:26] <fuzzie> that is ok, but shoulcommit fix this one? :)
[20:23:30] <fuzzie> ok, i can';t type
[20:23:32] <Avenger> it seems the changeanimation isn't always called, the troll seems to be confused
[20:23:55] <fuzzie> it sounded to me suspiciously like it was getting interrupted
[20:24:33] <fuzzie> maybe by some combat script
[20:24:34] <Avenger> hmm, it doesn't work
[20:24:39] <Avenger> even the changeanimation patch
[20:24:46] <Avenger> the troll is still at max hp :(
[20:25:18] <Akileos> Well, no can't reproduce, but not getting equipment either
[20:25:59] <fuzzie> ok. then i don't commit either change. :P
[20:28:04] <fuzzie> that is quite strange since you'd think the revert is fairly simple
[20:31:16] <lynxlynxlynx> uhh
[20:31:43] <lynxlynxlynx> why are irc clients so dumb? 3 people are talking and it colors two of them with the same color
[20:32:22] <lynxlynxlynx> i'll go check the equipment now
[20:32:45] <Avenger> ahh i was wrong, the change is good
[20:33:23] <Avenger> the troll seems to be 'uninjured' because it is also casting a 'troll death' spell on itself (spin700) it sets its max/current hps to 1
[20:33:44] <Avenger> which means, the change has no overall effect, but it is in the original :)
[20:33:55] <Avenger> so, the troll bug is still there
[20:34:11] <fuzzie> ok. this is terrible, it will turn out to be some stupid action bug
[20:34:20] <Avenger> yes, it seems
[20:34:31] <fuzzie> and there is so much broken there
[20:34:49] <fuzzie> i think an actionoverride would break it for example
[20:35:05] <lynxlynxlynx> the revert fixes it
[20:35:52] <fuzzie> ok. i'll commit both?
[20:35:57] <Akileos> And also, tried to create an entropist mage, but can do, stuck at speel selection, got to choose two in ... 0
[20:36:19] <CIA-27> GemRB: 03lynxlupodian * r2d27093f36b7 10gemrb/gemrb/GUIScripts/bg2/CharGenEnd.py:
[20:36:19] <CIA-27> GemRB: Revert "tob: don't give the starting gear twice"
[20:36:19] <CIA-27> GemRB: not needed anymore since a later commit fixed the core issue
[20:36:19] <CIA-27> GemRB: This reverts commit 62dd9fee407ed42b5ebf6013cf31f8503af52c9d.
[20:36:23] <-- Maighstir has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[20:36:23] <fuzzie> or not :)
[20:36:58] <lynxlynxlynx> too slow :P
[20:37:07] <fuzzie> i hate sf's git servers
[20:37:41] <fuzzie> which is odd because they're vastly superior to the svn ones
[20:37:42] <lynxlynxlynx> Akileos: huh, i can't either
[20:37:58] <lynxlynxlynx> another regression
[20:37:59] <fuzzie> hm i guess changing the commit message would've been smart there
[20:38:00] <CIA-27> GemRB: 03fuzzie * r6f59ae5ef852 10gemrb/gemrb/core/GameScript/GSUtils.cpp: Copy HP in ChangeAnimation, to fix magical healing trolls. (fix from Avenger)
[20:38:04] <fuzzie> see, this is why i shouldn't commit things
[20:38:19] <fuzzie> i am much better just mumbling things in the corner
[20:38:31] <Akileos> lynxlynxlynx: Sorry for finding bugs ;)
[20:38:42] <lynxlynxlynx> that's good, nothing to be sorry about
[20:38:50] <Akileos> No unit tests ?
[20:38:56] <fuzzie> it is good that someone is checking
[20:39:07] <lynxlynxlynx> none
[20:39:07] <fuzzie> Akileos: if you wrote the unit tests, you did 90% of the work of coding gemrb :(
[20:40:06] <fuzzie> it would be nice to have unit tests for the system/importer stuff, but that usually isn't where any bugs are
[20:40:07] <Akileos> What's actually missing ?
[20:40:14] <lynxlynxlynx> i don't remember any changes in that or related code though :/
[20:40:28] <lynxlynxlynx> it is clear at least where the surface problem is though
[20:40:47] <Akileos> Spell by level map ?
[20:41:43] <Avenger> hmm, weird, i have brought the troll down to 1 hp, and it doesn't call the animation change
[20:41:59] <fuzzie> well, the script does all kinds of weird stuff?
[20:43:02] <fuzzie> wait, it's calling *ChangeAnimation* and we replace the whole actor?
[20:43:04] <fuzzie> which script is this?
[20:43:40] <Avenger> trol*01
[20:43:48] <Avenger> any troll scripts
[20:43:58] <fuzzie> ok
[20:44:00] <fuzzie> so which game?
[20:44:02] <Avenger> changeanimation should replace the whole actor :)
[20:44:07] <Avenger> soa or tob
[20:44:21] <lynxlynxlynx> Akileos: how proficient are you with git?
[20:44:22] <fuzzie> yeah, i was just confused because the bug was in how, which uses PlayDead for trolls
[20:44:23] <Avenger> changeanimation replaces the whole actor, except orientation and hp
[20:44:34] <Akileos> lynxlynxlynx: 42
[20:44:51] <Avenger> i got playdead in the script too
[20:44:53] <Akileos> Depends on what you want me to do
[20:45:04] <lynxlynxlynx> Akileos: ok, can you try to bisect to the problematic commit then? :)
[20:45:12] <fuzzie> Avenger: how doesn't use the timers
[20:45:25] <Akileos> hahaha, for the sorcerer issue ?
[20:45:31] <lynxlynxlynx> yep
[20:45:45] <Akileos> Any idea where i should start ?
[20:45:59] --> Maighstir has joined #GemRb
[20:46:02] <Avenger> fuzzie: it doesn't even evaluate the triggers
[20:46:10] <lynxlynxlynx> try half a year ago
[20:46:39] <fuzzie> Avenger: maybe some non-interruptible attack is running?
[20:46:52] <Akileos> lynxlynxlynx: Really ?
[20:47:09] <lynxlynxlynx> sure, why not?
[20:47:25] <lynxlynxlynx> you need a good revision
[20:47:37] <lynxlynxlynx> bisection quickly closes the gap
[20:47:50] <Avenger> fuzzie: attackreevaluate or attack (the troll uses standard wtasight script for attack)
[20:48:08] <lynxlynxlynx> well, start where you want, but you'll need to find a working revision ;)
[20:48:33] <Avenger> the problem is, the troll is paralyzed, i guess we don't abort the attack spell when paralyzed?
[20:48:37] <fuzzie> Avenger: looks like someone made MoveNearerTo use IF_NOINT?!
[20:48:42] <Avenger> err attack action
[20:48:47] <Avenger> heh
[20:48:54] <Avenger> it cannot move anyway
[20:49:01] <fuzzie> i thought that was an iwd2 bug
[20:49:41] <fuzzie> you have this REed?
[20:49:41] <Avenger> btw, this is a separate bug
[20:49:49] <fuzzie> sure, i just don't want to forget about this
[20:50:10] <Avenger> i know what's wrong with movenearer+no interrupt
[20:50:24] <Avenger> that we already talked about: the original has no single bit
[20:50:38] <Avenger> it flags actions with the no interrupt bit when it queues them
[20:50:40] <fuzzie> i;m pretty sure i tried it in original engine and movenearer was always interruptible unless in iwd2
[20:50:45] <Avenger> each action has its own IF_NOINT
[20:50:49] <Avenger> it is not a global thing
[20:50:50] <fuzzie> but it has been a while
[20:51:01] <fuzzie> well, yes, but you made it set IF_NOINT, not the action flag :)
[20:51:21] <Avenger> well, since we don't have any action flag :)
[20:51:23] <Avenger> do we?
[20:51:24] <fuzzie> sure we do
[20:51:27] <Avenger> oh
[20:51:35] <fuzzie> well, we don't
[20:51:38] <Avenger> :P
[20:51:54] <fuzzie> but we have a flag on the actor which means 'this action can't be interrupted'
[20:52:05] <Avenger> that is IF_NOINT?
[20:52:07] <fuzzie> no
[20:52:14] <fuzzie> IF_NOINT is NoInterrupt flag
[20:52:38] <Avenger> hehe, i'm confused then
[20:52:40] <fuzzie> CurrentActionInterruptable is bool for 'don't interrupt action'
[20:53:01] <fuzzie> i'm pretty sure you have to set it to true every tick, though
[20:53:18] <Avenger> well...
[20:53:19] <fuzzie> so it has to go at a higher level of MoveNearerTo
[20:53:28] <fuzzie> so this is not so helpful, because i can't see what the original engine does
[20:53:43] <fuzzie> it definitely doesn't set the whole action bit here, you can interrupt chasing.
[20:53:59] <Avenger> i don't think this troll will ever move
[20:54:02] <fuzzie> except maybe for iwd2 which is insane and confusing and scary.
[20:54:11] <Avenger> it looks like the hold effect stops effect updates....
[20:54:15] <fuzzie> well, if the troll tried moving and had IF_NOINT set on it, then it has IF_NOINT set forever
[20:54:28] <Avenger> but it doesn't even move
[20:54:31] <Avenger> it is frozen
[20:54:42] <fuzzie> and it was frozen since it started existing?
[20:55:12] <Avenger> no
[20:55:19] <Avenger> i cast a chromatic orb on it :)
[20:55:23] <fuzzie> sure
[20:55:29] <fuzzie> so maybe it walked first?
[20:55:35] <Avenger> sure it did
[20:55:52] <Avenger> well, i'm not entirely sure if it walked or attacked
[20:55:57] <fuzzie> so this code just sets IF_NOINT and doesn't unset it
[20:55:59] <Avenger> the animation seems to be more like attack
[20:56:32] <Avenger> ok, i just broke this poor troll, it became statue
[20:56:55] <fuzzie> so it seems that most obvious explanation is that it walked and got IF_NOINT set and then got frozen and never runs scripts again :)
[20:57:27] <Avenger> if noint has effect on scripts?
[20:57:29] <fuzzie> and this makes sense because the bug report was about the troll dying as soon as the troll's last target died
[20:57:40] <fuzzie> IF_NOINT is *only* about scripts
[20:57:49] <Avenger> but running them?
[20:57:51] <Avenger> oh
[20:58:11] <fuzzie> oh, i see, it's used in dialog code too
[20:58:14] <fuzzie> ok, so i have no clue any more
[20:58:29] <Avenger> well, if the actor is held, it should probably abort its current action
[20:58:46] <Avenger> at least, it should abort moving, and casting and attacking and such
[20:58:50] <fuzzie> well
[20:58:59] <fuzzie> i'm not sure how this works
[20:59:04] <fuzzie> this is just the actual Hold effect?
[20:59:05] <Avenger> sadly this needs to be re-d first
[20:59:10] <fuzzie> ok
[20:59:22] <Avenger> yes, it is a hold effect, but the effect isn't on the troll by now
[20:59:24] <Avenger> it expired
[20:59:40] <Avenger> but the if noint flag may have stopped its script reevaluations
[20:59:46] <Avenger> so it won't queue any new action
[20:59:50] <fuzzie> that is my theory
[21:00:00] <Avenger> still don't know why it is not animated, though
[21:00:08] <Avenger> the hold effect isn't on it
[21:00:19] <Avenger> so i guess, it stopped reevaluating effects too
[21:00:25] <fuzzie> and that should not happen
[21:03:11] <Avenger> ok, it is regenerating, so the effects are running
[21:03:20] <Avenger> i don't understand why it isn't animated then
[21:03:28] <Avenger> the hold effect isn't on it
[21:04:17] <fuzzie> no such anim?
[21:05:01] <lynxlynxlynx> switching stances/anims wildly also causes a similar visual
[21:05:24] <Akileos> Jarvis:~/Devel/gemrb > git bisect good
[21:05:24] <Akileos> Bisecting: 55 revisions left to test after this (roughly 6 steps)
[21:05:24] <Akileos> [31c9f7a4b03aa631f2d3f76f2ee835ad8cb775a9] safer InCutSceneMode
[21:05:26] <lynxlynxlynx> it was an obvious before we fixed polymorph
[21:05:38] <lynxlynxlynx> :)
[21:05:55] <fuzzie> well, you get no anim update if you're running an action, i guess
[21:06:37] <Akileos> Wish I could understand the code to make unit tests :(
[21:06:41] <fuzzie> i would try just removing the IF_NOINT
[21:09:36] <Akileos> I really like the commit I'm at
[21:09:37] <Akileos> Bisecting: 27 revisions left to test after this (roughly 5 steps)
[21:09:38] <Akileos> [fb250e7bb828e1ba6c1720e8dc1af6edcdae5f4e] fix leaking Spell refcounts
[21:10:43] <Avenger> huh
[21:12:23] <Avenger> fuzzie: the troll is petrified :)
[21:12:34] <Avenger> so it is really a statue, LOL
[21:12:36] <fuzzie> :P
[21:12:50] <fuzzie> maybe not cast random things at the poor troll when testing
[21:12:57] <Avenger> ok
[21:15:27] <Avenger> lol., the fireball has no effect
[21:17:18] <lynxlynxlynx> i noticed the wand of fire didn't work in bg1 yesterday
[21:18:12] <lynxlynxlynx> i can confirm the bg2 fireball is just the same
[21:19:35] <lynxlynxlynx> too much pst projectile hackery?
[21:21:41] <Avenger> probably
[21:22:25] <Avenger> i don't know what's the problem, because the holy smite area effect works
[21:23:59] <Avenger> fireball has 0 explosion projectile... aha!
[21:24:34] <fuzzie> oh maybe this is the saradush bug
[21:24:53] <Avenger> don't think so
[21:25:03] <Avenger> the holy smite has inv travel projectile,
[21:25:31] <Avenger> but it could also be my doing
[21:25:33] <fuzzie> it was crashing due to explosion projectile being broken?
[21:25:38] <fuzzie> i mean, saradush
[21:25:54] <Avenger> well that red cloud is? storm of vengeance i guess
[21:26:05] <fuzzie> yes
[21:26:14] <pupnik_> n900 screen is too small for gemrb
[21:26:24] <Akileos> Does any of you have the orignal bg2.exe to compare ?
[21:26:27] <Avenger> it has the same inv travel as pholyst (the holy smite)
[21:26:51] <fuzzie> which has some other issue because areapro.2da has SPSRMVB under RESOURCE2 for VENGEANCE and that doesn't seem to exist
[21:27:01] <lynxlynxlynx> Avenger: the fact that we now use the storm is a bug in itself, it should be a comet like before
[21:27:04] <Avenger> oh
[21:27:08] <Avenger> a typo?
[21:27:09] <lynxlynxlynx> (mimicing catapults)
[21:27:11] <fuzzie> which brings me back to my question about why FIRIMP is in damage.2da and not SPFIRIMP
[21:27:28] <lynxlynxlynx> i fixed that
[21:27:33] <lynxlynxlynx> just not pushed
[21:27:37] <fuzzie> ok
[21:27:45] <lynxlynxlynx> the hardcoded defaults in actor.cpp already have spfirimp
[21:27:49] <fuzzie> i don't mean to act stupid, i just have no clue how this stuff works, so i am not going to touch it myself
[21:27:55] <lynxlynxlynx> and atleast for bg2, there is no such bam
[21:28:38] <Avenger> there is spstrmvb (missing t)
[21:28:59] <Avenger> anyway, it shouldn't crash on a missing resource
[21:29:12] <fuzzie> no, the missing resource is just something i noticed
[21:29:31] <Avenger> lynx, please fix that too
[21:29:33] <fuzzie> lynx tracked down the crash enough already
[21:29:45] <lynxlynxlynx> ok
[21:29:49] <Avenger> huh a comet?
[21:30:00] <Akileos> lynxlynxlynx: Private chat please
[21:30:06] <lynxlynxlynx> :P
[21:31:02] <Akileos> http://pastebin.com/4vhSqqk9
[21:31:31] <Akileos> Dont even understand how this coul be the cause
[21:31:52] <Avenger> the comet projectile (for catapults) is 'comet.pro'
[21:32:05] <Avenger> it doesn't use any fireball style animation
[21:32:09] <lynxlynxlynx> Akileos: sounds unlikely, but you never know
[21:32:13] <fuzzie> the autodetection thing moved the place where the python is inited, i think
[21:32:17] <lynxlynxlynx> if you revert that, does it go away?
[21:32:31] <Akileos> lynxlynxlynx: from master ?
[21:32:40] <lynxlynxlynx> i'll try if you don't want to, but i can do only three things at once ;)
[21:32:47] <lynxlynxlynx> yes, from master
[21:32:51] <lynxlynxlynx> HEAD i mean
[21:32:57] <lynxlynxlynx> well origin HEAD
[21:33:02] <lynxlynxlynx> :S
[21:33:07] <Avenger> ahh the comet starts another projectile: redhit, that has the dragonfire fireball effect
[21:33:13] <Avenger> 0xf
[21:33:30] <Avenger> 0xf in areapro.2da
[21:33:51] <Akileos> Jarvis:~/Devel/gemrb/build/gemrb > git revert fc6781a8c3b448c3fba4617139c3586391e33117
[21:33:52] <Akileos> error: could not revert fc6781a... Autodetect GameType if it is set to "auto"
[21:33:52] <Akileos> hint: after resolving the conflicts, mark the corrected paths
[21:33:52] <Akileos> hint: with 'git add <paths>' or 'git rm <paths>'
[21:34:15] <Akileos> don't want to mess up everything
[21:34:16] <Avenger> that should be this-->DRAGON SPRDRASI * * EFF_M21 * 16
[21:34:56] <Avenger> and there is no such anim ;)
[21:35:08] <Avenger> ah there it is
[21:35:25] <lynxlynxlynx> sure, i already saw it from gemrb at various times
[21:35:27] <CIA-27> GemRB: 03lynxlupodian * r584ec9bf4bb6 10gemrb/gemrb/override/bg2/damage.2da: bg2/damage.2da: firimp doesn't exist, but spfirimp does
[21:35:32] <lynxlynxlynx> Akileos: maybe you have a dirty tree
[21:35:34] <CIA-27> GemRB: 03lynxlupodian * re129568460dd 10gemrb/gemrb/override/ (bg2/areapro.2da how/areapro.2da): areapro.2da: fixed SPSTRMVB typos
[21:35:45] <lynxlynxlynx> i'll go try it
[21:36:13] <Avenger> so i still don't know what's up with the comet projectile, it worked before
[21:37:05] <lynxlynxlynx> ah no, it's a bad merge here too
[21:37:32] <lynxlynxlynx> Avenger: probably nothing, i checked the history the other day and comet.pro hasn't been modified since the import
[21:38:23] <Akileos> Ok guys, got to go to bed I I want to earn my paycheck tomorrow
[21:39:29] --- Akileos is now known as Akileos-ZzZ
[21:39:46] <Akileos-ZzZ> lynxlynxlynx: do hesitate to PM me
[21:40:12] <lynxlynxlynx> uh ok
[21:40:37] * lynxlynxlynx hesitates
[21:42:11] --> |Cable| has joined #GemRb
[21:42:30] <-- Avenger has left IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726])
[21:49:12] <-- kettuz has left IRC (Quit: Leaving)
[22:11:26] <edheldil_> now I do not know whether itreally was my commit which broke it :)
[22:12:15] <fuzzie> well, you changed a bunch of the python setup bits in one commit
[22:12:29] <fuzzie> so, maybe
[22:14:13] <edheldil_> sure
[22:15:03] <fuzzie> i couldn't see any relevant changes at all
[22:15:08] <fuzzie> so it would be some weird change, i guess
[22:21:03] <fuzzie> oh hm
[22:21:18] <fuzzie> also you are sprintfing into a pretty tiny buffer
[22:21:44] <fuzzie> that was there before though
[22:22:18] <fuzzie> but i wouldn't be surprised if someone makes a path exceeding ~200 chars someday
[22:25:46] <fuzzie> oh i see the change
[22:26:13] <fuzzie> you broke the override
[22:27:58] <fuzzie> which i guess explains all our bugs
[22:29:48] <edheldil_> hmm?
[22:30:09] <fuzzie> you add the override to the search path now *after* everything else
[22:30:15] <fuzzie> so it doesn't override anything, so all the gemrb data is ignored :P
[22:30:47] <edheldil_> do I? Checking ...
[22:31:00] <fuzzie> i am going to simply break HasResource for your autodetection for now, and move the guiscript init before the search path code
[22:31:10] <fuzzie> unless you think you can come up with a real fix now
[22:34:51] <lynxlynxlynx> oh yeah, i thought that was odd too
[22:35:11] <lynxlynxlynx> need to be more like fuzzie and complain asap
[22:35:51] <fuzzie> i did check the diff, i didn't notice
[22:36:05] <lynxlynxlynx> iirc it was a separate commit in the fork
[22:36:22] <fuzzie> i don't know how to check chargen though
[22:36:39] <fuzzie> because i do't understand the bug report :)
[22:38:28] <lynxlynxlynx> hmm, nope
[22:38:53] <lynxlynxlynx> if you can create a wild mage then it's working again
[22:39:36] <fuzzie> then yes, it's fine
[22:39:58] <fuzzie> i see that 'pick for me' doesn't pick the mandatory one though
[22:40:30] <fuzzie> initial equipment is working fine also.
[22:40:41] <fuzzie> i will wait for edheldil's reaction before i commit this though
[22:41:15] <fuzzie> i'm not sure if we need the HasResource thing :/
[22:41:57] <fuzzie> but the alternative is bashing a hole in the ResourceManager interface to let you insert at arbitary points, i guess, and that is horrible
[22:41:59] <edheldil_> Hmm,somehow I forgot that problem
[22:42:23] <fuzzie> so if this works without HasResource for now, it is quick simple temporary solution
[22:43:33] <fuzzie> otherwise say 'don't commit' :)
[22:43:52] <edheldil_> Does it really work? iirc it's used ...
[22:44:16] <fuzzie> i don't know, i don't have unmodified installs
[22:44:57] <fuzzie> but i doubt it, since you don't even check override in AutodetectCommon
[22:46:37] <edheldil_> ah, sure, I just check an existence, which should be in chitin.key
[22:47:03] <fuzzie> but then if i commit this, i break your autodetection for now :/
[22:49:29] <edheldil_> well, commit it, I will check. I think that inserting paths is not so bad, I do it already for GUIScripts paths
[22:50:00] <edheldil_> I am too fuzzy now to think about it
[22:50:03] <fuzzie> ok
[22:50:07] <fuzzie> me too, really
[22:50:15] <fuzzie> we fix it tomorrow or weekend or something :)
[22:52:24] <CIA-27> GemRB: 03fuzzie * rf851e2edc9c0 10gemrb/gemrb/core/Interface.cpp: Fix search path order. This probably breaks autodetection for now.
[22:52:25] <fuzzie> i am fine with whatever works i just don't want to do big changes at 10 minutes before midnight
[22:54:22] <lynxlynxlynx> some spell levels don't have any specialist spells in the recommended list for some schools
[22:54:51] <-- Maighstir has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[22:54:58] <lynxlynxlynx> considering wild mages were added last and have unique spells, it is no surprise those aren't on the common suggestion list
[22:55:33] <lynxlynxlynx> trying out the commit
[22:57:27] <lynxlynxlynx> ok, the comet issue is fixed
[22:57:38] <lynxlynxlynx> the fireball one isn't
[22:58:03] <fuzzie> drat.
[22:59:07] <lynxlynxlynx> i can confirm the wild mages are fixed too
[23:03:40] <fuzzie> well i guess too much spell/projectile stuff changed recently for me to understand
[23:03:41] <-- lynxlynxlynx has left IRC (Read error: Connection reset by peer)
[23:04:31] <fuzzie> so sleepy time
[23:34:18] <-- Bo_Thomsen has left IRC (Quit: Leaving.)