#gemrb@irc.freenode.net logs for 14 Apr 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:06:39] <-- tomprince has left IRC (Remote host closed the connection)
[00:07:06] --> tomprince has joined #gemrb
[00:15:20] <-- tomprince has left IRC (Remote host closed the connection)
[00:15:45] --> tomprince has joined #gemrb
[01:11:03] --> raevol has joined #gemrb
[01:35:31] <-- Demitar has left IRC (Quit: Burn the land and boil the sea, you can't take the sky from me!)
[01:47:34] <-- raevol has left IRC (Quit: Leaving.)
[01:50:30] --> PixelScum has joined #gemrb
[01:53:58] <-- BaldimerBrandybo has left IRC (Ping timeout: 260 seconds)
[05:23:41] --> ubermad has joined #gemrb
[06:25:21] --> Bo_Thomsen has joined #gemrb
[06:52:09] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[07:12:08] --> lynxlynxlynx has joined #gemrb
[07:12:08] --- ChanServ gives channel operator status to lynxlynxlynx
[07:19:09] --> adominguez has joined #gemrb
[09:08:32] <-- Cable_ has left IRC (Remote host closed the connection)
[10:02:18] <edheldil> regarding SDLVideo, what about splitting it to several files? E.g. input event marshalling / blitter / sprite transformations / video init, fullscreen
[10:02:38] <edheldil> at least as includes
[10:03:16] <fuzzie> we should make the event/etc stuff a superclass
[10:05:26] <fuzzie> because then we can share it with a sdl+opengl video class
[10:05:41] <fuzzie> but i think the rest of it is probably best how it is
[10:05:47] <fuzzie> it's all interlinked anyway
[10:06:02] <edheldil> I haven't looked at android hacks, but pinch surely needs something in input as well. Maybe the android subclass could extend it. Even so, would splitting the file make it more manageable?
[10:06:34] <fuzzie> i mean, the video/blitter stuff is all interlinked
[10:09:43] <edheldil> not THAT much, I think - e.g. both flipsprite and blitsprite need to know pixel layout, of course, but they do not really depend on each other
[10:10:31] <edheldil> nevermind, if you think it's a bad idea, that's fine with me :)
[10:10:53] <fuzzie> well, i mean more that the MirrorSprite functions just set booleans that the blitter obeys, for BAMs
[10:13:03] <fuzzie> AddPolygonToSpriteCover being in the video driver is a bit ridiculous
[10:13:51] <wjp> a few parts could be extracted upward (especially main loop/event handling related, I think), but I don't think there's really anything to split up
[10:14:01] <fuzzie> given that SpriteCover is pretty software-renderer-specific
[10:14:47] <wjp> hm, no, it can be done with scissors-magic in opengl too
[10:15:32] <fuzzie> with a pixel buffer?
[10:16:05] <wjp> maybe I misunderstand what you meant
[10:16:26] <fuzzie> right now, the core has a 'SpriteCover' class, containing pixels
[10:16:45] <fuzzie> and it calls into the video driver to convert the polygon to that form
[10:17:22] <wjp> oh, that's a bit weird, yes
[10:18:24] <wjp> I can only assume that opengl wasn't being considered yet when I wrote that :-)
[10:18:28] <fuzzie> I mean, we *could* draw that as a texture to make a stencil, but seems a little insane. ;p
[10:18:57] <wjp> my opengl experiments ages ago used a static checkerboard stencil, and then scissored out the polygons
[10:19:35] <fuzzie> stenciled, or actually scissored?
[10:19:39] <wjp> I kind of hope there are more elegant ways, but I don't really know opengl at all, let alone recent versions
[10:19:53] <fuzzie> yes, well, me neither :)
[10:20:26] <wjp> looking at the code there is no actual scissoring involved
[10:20:27] <fuzzie> but you're the one who works with graphics cards all day, you should be our expert at how they work, right? :)
[10:20:43] <wjp> just drawing the covered polygons with a stencil test and some depth checking
[10:21:08] <wjp> did I mention some of my graphics cards don't have any video out at all :-)
[10:22:06] <wjp> I think the reason for SpriteCover being in the place it is, is that there used to be two bitmap-based renderers in the past (SDL and DirectX)
[10:22:27] <fuzzie> ah
[10:22:30] <wjp> I should also add that the SpriteCover stuff is the very first code I ever wrote for gemrb
[10:22:55] <fuzzie> hehe!
[10:23:41] <fuzzie> my first commit was a terrible, terrible hack which happens to be exactly the right code in the right place
[10:23:46] <wjp> :-)
[10:23:48] <wjp> nice
[10:23:54] <fuzzie> which i only discovered a couple of weeks ago
[10:25:54] <lynxlynxlynx> :)
[10:26:38] <lynxlynxlynx> i think i started with the mage spell learning in bg2 cg
[10:26:54] <lynxlynxlynx> buttons buttons buttons
[10:39:25] --> spike411 has joined #gemrb
[10:54:01] --> Maighstir has joined #gemrb
[11:03:37] <-- spike411 has left IRC (Ping timeout: 246 seconds)
[11:12:03] --> SiENcE has joined #gemrb
[11:13:38] --> Drakkar has joined #gemrb
[11:15:53] <-- PixelScum has left IRC (Ping timeout: 248 seconds)
[11:33:38] --> spike411 has joined #gemrb
[11:33:42] <-- SiENcE has left IRC (Ping timeout: 246 seconds)
[11:47:44] <edheldil> I started by hacking on pst guiscripts, iirc
[12:59:00] --> test32894789234u has joined #gemrb
[14:42:31] <-- spike411 has left IRC (Quit: Manga & anime pokec na Jabberu: manga.cz@conf.netlab.cz)
[14:57:01] <-- Gekz has left IRC (Read error: Connection reset by peer)
[14:58:55] --> Gekz has joined #gemrb
[15:03:22] <-- Gekz has left IRC (Ping timeout: 246 seconds)
[15:11:58] --> Gekz has joined #gemrb
[15:54:02] <-- lubos has left IRC (Quit: Leaving.)
[16:20:23] <-- edheldil has left IRC (Remote host closed the connection)
[16:47:58] --> edheldil has joined #gemrb
[16:47:58] --- ChanServ gives channel operator status to edheldil
[17:27:40] <dhewg> whats "[Actor]: Unregistered attack. We shouldn't be here?" about?
[17:28:07] <dhewg> sometimes i get that alot on fights involving lots of enemys, but sometimes not
[17:28:50] <fuzzie> fixed on master :P
[17:29:29] <fuzzie> previously we had this complicated system involving an attackers array
[17:29:48] <dhewg> cherry-pickable?
[17:29:51] <fuzzie> i never quite worked out the point, but removing all the code seemed to work well
[17:29:58] <dhewg> heh :)
[17:29:59] --> |Cable| has joined #gemrb
[17:30:19] <fuzzie> not sure if cherry-picking it is advisable :P
[17:31:06] <dhewg> hmpf
[17:31:20] <dhewg> whats the status of your local changes? :D
[17:31:35] <dhewg> i mean, surely it makes more sense to report stuff when using master
[17:31:46] <fuzzie> sword25 renders correctly now! i mean, wait, gemrb. i don't know, disorganised.
[17:31:58] <dhewg> moar gemrb!
[17:32:00] <fuzzie> last time i tried everything was still dying again every frame.
[17:32:15] <dhewg> that doesnt sound like a fix to me :P
[17:32:48] <fuzzie> i mean, in my local changes
[17:32:58] <dhewg> yeah, i know
[17:33:02] <dhewg> but now that you mentioned it, i got that once too on the branch
[17:33:14] <dhewg> it was before i used that case fix
[17:33:24] <fuzzie> we have some really weird setup in the main tree
[17:33:43] <fuzzie> where things sort of half-die and then we wait for stuff to happen before they die
[17:34:05] <dhewg> remember the room on the 1st dungeon with the respawning mephits?
[17:34:25] <dhewg> where you have to kill the spawny cages
[17:34:30] <fuzzie> i hate the trigger in that room :)
[17:34:34] <dhewg> :)
[17:34:52] <fuzzie> if it's a complaint about the trigger, i made that one worse too ;p
[17:35:09] <dhewg> well, as i removed that room file in my cache while the game was playing, and reentered, i got that effect that everything was dropping dead like 10 times a second
[17:35:25] <dhewg> dunno, but maybe that helps to get an idea?
[17:35:32] <fuzzie> that's interesting
[17:37:01] <dhewg> lots of dead mephits too
[17:37:52] <dhewg> or maybe i ran rm * in the cache dir, cant remember
[17:40:00] <tomprince> You don't want to do 'rm *'.
[17:40:15] <fuzzie> well you don't want to do anything now :-)
[17:40:29] <tomprince> rm'ing the area you are currently in works, since when you leave, the area gets saved.
[17:40:44] <dhewg> i did that `rm *` because the case bug broke my quests, now i do not do that :)
[17:40:59] <tomprince> rm * would cause any area that you have been to so far in this game to be reload from the original state.
[17:41:26] <fuzzie> i'm not sure what would cause death stuff
[17:41:33] <fuzzie> but i guess i have to rewrite all that coode too, bleh
[17:41:41] <dhewg> fun!
[17:43:35] <fuzzie> well maybe i can sucker someone else into doing it..? :-)
[17:45:18] <dhewg> heh
[17:46:46] <dhewg> damnit, aerie just started her whining during a fight
[17:47:04] <fuzzie> is that meant to happen?
[17:47:54] <dhewg> cant remember, but that looks like its not supposed to happen
[17:48:17] <fuzzie> it looks like some of the introductory love talks don't have a CombatCounter check in the vanilla game
[17:48:58] <tomprince> fuzzie: The only problem with getting somebody else to rewrite all the stuff you want to rewrite, is that is seems that what needs to be implemented exists only in your head.
[17:49:15] <fuzzie> i am much better at babbling on irc about it than i am at actually coding it!
[17:49:16] <tomprince> So, all you need to do is figure out how to do a brain transplant first. :)
[17:50:14] <fuzzie> i wrote http://fuzzie.org/nfs/gemrb/func_movetoobject.txt with the idea of writing more like that
[17:50:32] <fuzzie> but then i realised that it doesn't really help, because that *depends* on a bunch of stuff which is only in my head
[17:54:18] <tomprince> but if you keep writting stuff like that, eventually you will get to stuff that doesn't depend on stuff in your head.
[17:55:19] <dhewg> which ruins the initial plan to make someone else do it
[17:56:16] <tomprince> not entirely, since that is at a (atleast slightly) higher level than actual gemrb code.
[17:56:50] <tomprince> and perhaps might lead to figuring out how to make actual gemrb code able to written at that level.
[17:57:49] <tomprince> Looking at that code, i feel almost tempted to say that it should be implemented in python
[18:00:10] <tomprince> (that may just be because I have been staring at a bunch of python and twisted code recently)
[18:00:39] <fuzzie> being able to yield in some way makes everything nicer :/
[18:02:24] <-- |Cable| has left IRC (Remote host closed the connection)
[18:04:00] --> |Cable| has joined #gemrb
[18:05:51] <-- |Cable| has left IRC (Remote host closed the connection)
[18:07:13] --> |Cable| has joined #gemrb
[18:09:30] <-- ubermad has left #gemrb
[18:34:40] <-- adominguez has left IRC (Remote host closed the connection)
[18:46:44] --> Maighstir_ has joined #gemrb
[18:49:37] <-- Maighstir has left IRC (Ping timeout: 240 seconds)
[19:04:15] <-- test32894789234u has left #gemrb
[19:19:23] <lynxlynxlynx> progress by direct annoyance
[19:19:38] <lynxlynxlynx> merge your changes and i'm sure we'll find out why things die all the time there asap
[19:33:35] <dhewg> i just ran into the shop issue again
[19:34:08] <dhewg> in the tavern in the slums, i rented a room, but it was interrupted by jaheira having a nightmare
[19:34:26] <dhewg> after that bernhard doesnt show the sales screen anymore
[19:34:38] <dhewg> reloading doesnt help either, restarting does
[19:34:43] <fuzzie> yes, once the GUI breaks it's broken for the session
[19:35:03] <fuzzie> We should, again, be paused in stores.
[19:35:22] <lynxlynxlynx> i'll go fix that right now
[19:35:29] <lynxlynxlynx> stores are nice, i already merged them
[19:35:41] <fuzzie> well
[19:35:46] <fuzzie> sorry, i don't mean that, that is a random guess :-P
[19:35:49] <dhewg> i have a savegame to reproduce
[19:35:50] <fuzzie> but it sounds like a good answer
[19:35:59] <dhewg> does that help?
[19:36:13] <dhewg> she has that dream right there on renting
[19:36:25] <fuzzie> but maybe it's more complicated than that, because, the player dream scenes are also sleep-triggered?
[19:36:45] <fuzzie> and the original engine might well return us to the store after those
[19:36:48] <dhewg> its not a cutscene dream, its just a dialog
[19:37:06] <fuzzie> yes, the dialog thing can be fixed by just making sure to pause in stores
[19:37:10] <fuzzie> just thinking, there's probably a bigger problem
[19:38:20] <dhewg> shouldnt the game pause on every fullscreen gui thing?
[19:38:25] <fuzzie> no
[19:38:34] <lynxlynxlynx> almost all
[19:38:37] <fuzzie> not in the inventory, and not in the world map
[19:38:47] <dhewg> i had a gui breakage on level up, on the store, on a dream and i bet there's more
[19:39:02] <fuzzie> but yes, everywhere else :-P
[19:39:21] <fuzzie> i don't know if we have a way to do a stacked non-feedback pause?
[19:41:15] <fuzzie> doesn't look like it
[19:42:11] <dhewg> the party was also not rested
[19:42:15] <dhewg> dunno if thats right
[19:42:44] <dhewg> and if the shop breaks saving doesnt work either anymore :P
[19:42:59] <fuzzie> 'q' works from pretty much everywhere i think
[19:43:01] <fuzzie> sigh, though
[19:43:26] <dhewg> how come the gui is so fragile?
[19:43:41] <dhewg> is it the architecture, because its driven by game scripts or something?
[19:44:18] <fuzzie> the python code has a system for closing the current window
[19:44:34] <fuzzie> there's just nothing hooked up to it for windows which aren't meant to be interruptible
[19:45:09] <fuzzie> on top of that, for cutscenes, the C++ code destroys the gui state without notifying the python
[19:45:38] <fuzzie> i moved the gui reset to the python side for dialogs a few weeks ago, just haven't done it for cutscenes yet.
[19:46:37] <dhewg> k, makes sense
[19:47:17] <dhewg> prolly a retarded question, but why's the gui seperated into py scripts in the first place?
[19:48:11] <lynxlynxlynx> easier to mod
[19:48:59] <fuzzie> and we're rather more lacking in C++ coders than in python coders
[19:49:00] <dhewg> for this or external projects?
[19:49:12] <fuzzie> i'm hoping to move more of the game logic into python at some point too..
[19:49:32] <fuzzie> but for external projects, the difficulty in modifying the original engine's hardcoded GUI is a common modder complaint
[19:49:45] <fuzzie> since people would like to rewire the whole thing
[19:50:32] <dhewg> yeah, while i dont know much about infinity, i already got the impression while looking at how the ws mod works
[19:51:05] <fuzzie> so with that in mind, 'easy to understand+modify for people who are inexperienced with coding' is possibly our primary goal for the GUI code
[19:51:22] <dhewg> k
[19:51:26] <fuzzie> i don't think we've managed that at all, mind :-/
[19:51:27] <dhewg> are there such projects? :)
[19:51:33] <lynxlynxlynx> more than 6 party members!!!1111 ;)
[19:52:07] <lynxlynxlynx> we already incorporated some gui tweaks, even some 3rd party ones
[19:52:07] <fuzzie> i think we ended up merging the only code anyone developed publically, with the guienhancements option
[19:52:31] <dhewg> hm
[19:52:51] <dhewg> on the other side, isnt python a little overkill - with small devices and portability in mind?
[19:52:55] <fuzzie> i think taimon promptly went and hacked those into the original engine
[19:53:17] <fuzzie> well, so far python has been the least problem anyone's had
[19:53:57] <fuzzie> you need a considerable amount of free disk space for the data cache
[19:54:05] <dhewg> heh right, with the android sdl port, python is not the primary issue i guess :)
[19:54:34] <fuzzie> and python isn't in any performance-sensitive paths, it's just an annoying dependency and some fairly small RAM usage
[19:55:11] <fuzzie> in any case we really don't have the manpower to mess with it too much now without driving everyone insane
[19:55:42] <dhewg> yeah
[19:56:05] <dhewg> dont get me wrong, im trying to get a feeling how this beast works
[19:56:17] <fuzzie> mostly, with duct tape :-)
[19:56:24] <dhewg> :D
[19:56:39] <fuzzie> tomprince has been doing some really nice refactoring work, in more recent times
[19:57:18] <fuzzie> but i am rather grumpily holding the fort against adding dependencies like boost::python, especially since there's no way our lead dev would make it work
[19:57:35] <fuzzie> so in the end it's all duct tape
[19:58:05] <dhewg> heh
[19:58:12] <dhewg> yeah, boost is weird
[19:58:26] <dhewg> while it has some really nice features, its a monster
[19:58:59] <dhewg> we use boost::python at work, and the part that includes that takes ages to compile because of its template voodoo
[19:59:58] <lynxlynxlynx> it's a distribution nightmare too
[20:00:09] <lynxlynxlynx> the abi changes much to often
[20:09:34] <dhewg> oh, i just got this on bg2, only saw it on bg1 before: [CharAnimations]: Couldn't load animation: mberg2e, cycle 22
[20:26:36] <lynxlynxlynx> same files
[20:41:32] <lynxlynxlynx> ups, got distracted
[20:46:27] <lynxlynxlynx> resting in a paused store causes problems
[20:46:46] <lynxlynxlynx> no cursor :=)
[20:47:41] <fuzzie> UpdateActionsWindow gets called in there
[20:48:01] <fuzzie> but no sure why you'd lose the cursor, unless a cutscene started
[20:48:46] <lynxlynxlynx> possibly
[20:49:00] <lynxlynxlynx> all i saw was the resting video though and there's no bcs lookup
[20:49:14] <fuzzie> so probably not, then
[20:49:23] <lynxlynxlynx> oh but there is
[20:50:07] <lynxlynxlynx> yep, a dream
[20:50:45] <fuzzie> that is .. a bit complicated to fix
[20:50:58] <fuzzie> due to me having broken stuff
[20:51:38] <lynxlynxlynx> pause doesn't stop scripts anymore
[20:51:48] <lynxlynxlynx> do we have a hard pause alternative?
[20:52:37] <fuzzie> i messed up the instants stuff, so you can't fix the script thing
[20:52:38] <lynxlynxlynx> ah, it does it only for dialogs
[20:53:08] <fuzzie> oh, dialogs shouldn't be instant
[20:53:34] <lynxlynxlynx> i'm not sure i should commit then, a broken store is better than a broken game
[20:53:59] <fuzzie> i think it's broken either way, if you can't save etc either
[20:54:28] <fuzzie> we should really have a way to do a stacked pause and fix the other screens, at least
[20:54:41] <lynxlynxlynx> quicksave works, but also outside resting
[20:55:00] <lynxlynxlynx> i added the pause thing on my todo
[20:55:04] <fuzzie> :)
[20:55:17] <lynxlynxlynx> it reminds of the merging horrors ahead
[20:55:59] <fuzzie> What's next, GUILOAD?
[20:55:59] <lynxlynxlynx> but a challenge is a challenge, we don't play rpgs because we're weak ;)
[20:56:37] <lynxlynxlynx> the most duplicated is guicommonwindows and some other windows
[20:56:59] <lynxlynxlynx> guiopt is another behemot
[20:57:27] <lynxlynxlynx> guiload is already so different i'm not sure it makes much sense in merging it
[20:57:37] <fuzzie> the commonwindows stuff is a bit complicated
[20:57:51] <fuzzie> i seem to remember it's pretty broken for iwd2/pst at least
[20:58:57] <dhewg> got this as i stacked 2 scrolls: http://pastie.org/1795647
[21:00:57] <lynxlynxlynx> yes, it will be easier to merge the others first
[21:01:24] <fuzzie> pastie.org dns isn't working for me
[21:02:28] <lynxlynxlynx> the new charge assert triggered
[21:03:28] <fuzzie> oh, it should be >=0
[21:04:00] <fuzzie> someone who is awake should fix that, sorry
[21:06:21] <lynxlynxlynx> ok
[21:06:40] <CIA-52> GemRB: 03lynxlupodian * re73b9d1097a0 10gemrb/gemrb/GUIScripts/bg2/GUILOAD.py: bg2/GUILOAD.py: don't set the expansion mode manually
[21:06:45] <CIA-52> GemRB: 03lynxlupodian * rdec1261b0b9a 10gemrb/gemrb/GUIScripts/GUISTORE.py: pause the game during store transactions
[21:06:51] <CIA-52> GemRB: 03lynxlupodian * rdbc9ce9c3d20 10gemrb/gemrb/ (core/Interface.cpp plugins/CHUImporter/CHUImporter.cpp): removed the useless "Cannot Load BackGround, skipping" message
[21:07:15] <fuzzie> our hero
[21:08:31] <lynxlynxlynx> how would >=0 help though, wouldn't the previous if block catch it?
[21:14:39] <lynxlynxlynx> ok, the hooks should also be snappier now
[21:17:43] <lynxlynxlynx> i can merge potions normally
[21:17:56] <lynxlynxlynx> dhewg: are you on master now?
[21:18:15] <dhewg> no, still 0.6.4 with a couple cherry-picks
[21:19:29] <lynxlynxlynx> can you get back into that situation?
[21:19:39] <lynxlynxlynx> i wonder why the chunk count got negative
[21:19:48] <dhewg> the assert?
[21:19:51] <lynxlynxlynx> yes
[21:19:59] <lynxlynxlynx> p myslot->Usages[0]
[21:20:01] <lynxlynxlynx> p myslot->StackAmount
[21:20:13] <lynxlynxlynx> p item->Usages[0]
[21:20:14] <dhewg> im not sure if i can reproduce
[21:20:36] <dhewg> i was shuffling inv stuff for like 5 minutes, with stacking, unstacking and writing magic :)
[21:22:07] <lynxlynxlynx> you said you stacked two scrolls - were they alone or were there more in the stack?
[21:23:14] <dhewg> i'll try, but i know that just stacking 2 scrolls worked before. maybe it was a sequence or something
[21:23:50] <lynxlynxlynx> looks like the assert should only be in the branch where it changes
[21:25:46] <dhewg> but i just cherry-picked your 3 commits there and tried to rent a room again
[21:26:10] <dhewg> i think a dialog or sequence is active, but i still see the sales screen
[21:26:17] <dhewg> while the game seems to be paused :)
[21:26:28] <lynxlynxlynx> you can exit it if you press ctrl+space
[21:26:37] <lynxlynxlynx> then type EndCutSceneMode()
[21:27:33] <dhewg> yay
[21:27:47] <dhewg> yeah, its jaheira's nightmare again
[21:28:47] <dhewg> but this banter behaves weird
[21:29:48] <dhewg> i got one of irenicus dream cutscenes, and before he said his first line, aerie starts wis her whining
[21:29:57] <dhewg> while the cutscene already started
[21:32:54] <lynxlynxlynx> i guess multiple things triggered
[21:35:03] <lynxlynxlynx> what did aerie say? give me a distinctive word to search for
[21:35:15] <dhewg> uhm
[21:35:20] <dhewg> sec
[21:38:02] <dhewg> "need you to tell me what the worth"
[21:38:32] <lynxlynxlynx> thanks
[21:39:20] <lynxlynxlynx> worth doesn't match anywhere :s
[21:39:45] <dhewg> it start with "I...need you"
[21:40:51] <fuzzie> lynxlynxlynx: you're right, that's weird
[21:41:12] <lynxlynxlynx> i suspect the data was already negative
[21:41:22] <lynxlynxlynx> dhewg: proper dialog and all?
[21:41:25] <fuzzie> yes, i guess we don't check if StackAmount > 1
[21:41:39] <lynxlynxlynx> i can't find anything similar even with just -i "need you"
[21:42:07] <lynxlynxlynx> oh, nevermind
[21:42:13] <lynxlynxlynx> banters are in other files
[21:42:17] <dhewg> :)
[21:42:58] <lynxlynxlynx> just two globals checked and !StateCheck(Player1,STATE_SLEEPING)
[21:43:27] <lynxlynxlynx> i don't think we can sneak a cutscene check in there
[21:43:45] <lynxlynxlynx> but that's fuzzie's area of expertise
[21:44:07] <fuzzie> it is easier to fix that at a later time
[21:52:48] --> Bo_Thomsen has joined #gemrb
[22:06:39] <CIA-52> GemRB: 03lynxlupodian * r5e782232036d 10gemrb/gemrb/core/Inventory.cpp:
[22:06:39] <CIA-52> GemRB: Inventory::AddSlotItem: added a guard for the rare occasion that the stack
[22:06:39] <CIA-52> GemRB: amount is not set properly (triggered unreliably with some scrolls)
[22:06:42] <CIA-52> GemRB: 03lynxlupodian * r850b52c54892 10gemrb/gemrb/core/Inventory.cpp: Inventory::AddSlotItem: use ASI_FAILED on failure
[22:10:14] <-- Maighstir_ has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[22:17:18] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[22:52:05] --> edheldil_ has joined #gemrb
[23:31:49] <-- Bo_Thomsen has left IRC (Quit: Leaving.)