#gemrb@irc.freenode.net logs for 10 Apr 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:43:24] --> _pickle has joined #gemrb
[01:03:41] <-- Maighstir has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[01:05:14] --> pupnik_ has joined #gemrb
[01:08:58] <-- pupnik has left IRC (Ping timeout: 248 seconds)
[01:25:31] <-- _pickle has left IRC (Remote host closed the connection)
[08:00:37] --> test32894789234u has joined #gemrb
[09:08:37] <pupnik_> o/
[09:37:35] <-- test32894789234u has left #gemrb
[09:37:46] --> test32894789234u has joined #gemrb
[10:13:35] <dhewg> bg1 segfaults on master when loading a savegame: http://pastie.org/1778297
[10:16:02] <dhewg> its also not possible to start a new game, when hitting 'accept' on the char creatin screen: http://pastie.org/1778300
[10:24:03] <pupnik_> hmm
[10:24:55] <pupnik_> version 0.6.4?
[10:25:16] <dhewg> both work fine with 0.6.4, those pasties are with master
[10:25:24] <dhewg> both compiled on linux 64bit
[11:14:09] --> Maighstir has joined #gemrb
[12:18:54] <-- test32894789234u has left #gemrb
[12:50:48] <tomprince> does somebody have a version of cmake <=2.6.4, or do we care about those versions?
[12:51:49] <pupnik_> cmake version 2.6-patch 2
[12:54:27] <tomprince> Can you you try running an in-tree build with both -DINSOURCEBUILD=Yes and -DINSOURCEBUILD=No with and without this patch? https://gist.github.com/912318
[13:08:03] <tomprince> (I'm running into issues with http://www.cmake.org/cmake/help/cmake-2-8-docs.html#policy:CMP0012 )
[13:14:22] <tomprince> dhweg: Can you do a 'print wpsattack' on the first one?
[13:29:01] --> PixelScum has joined #gemrb
[13:31:10] <pupnik_> ?
[13:31:47] <-- BaldimerBrandybo has left IRC (Ping timeout: 264 seconds)
[13:32:37] <dhewg> tomprince: $1 = (int **) 0xca9d00
[13:33:26] <dhewg> if it helps: http://pastie.org/1778802
[13:35:53] <tomprince> Well, the issue is clearly tmplevel = -1
[13:39:04] <tomprince> how about wpsattack_cols?, that is probably 0, which means that wspatck.2da is somehow wrong...
[14:14:38] --> Drakkar has joined #gemrb
[14:16:02] <-- PixelScum has left IRC (Ping timeout: 276 seconds)
[14:17:47] --> PixelScum has joined #gemrb
[14:20:08] <-- Drakkar has left IRC (Ping timeout: 246 seconds)
[14:47:57] --> kettuz has joined #gemrb
[14:51:31] <-- boriskr has left IRC (Read error: Operation timed out)
[14:52:39] --> boriskr has joined #gemrb
[15:04:17] --> Drakkar has joined #gemrb
[15:06:48] <-- PixelScum has left IRC (Ping timeout: 260 seconds)
[15:30:49] <-- spike411 has left IRC (Quit: Manga & anime pokec na Jabberu: manga.cz@conf.netlab.cz)
[15:38:33] --> PixelScum has joined #gemrb
[15:41:45] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[16:08:54] <-- PixelScum has left IRC (*.net *.split)
[16:08:54] <-- boriskr has left IRC (*.net *.split)
[16:08:55] <-- dhewg has left IRC (*.net *.split)
[16:10:56] --> PixelScum has joined #gemrb
[16:10:56] --> boriskr has joined #gemrb
[16:10:56] --> dhewg has joined #gemrb
[16:21:06] <-- boriskr has left IRC (Read error: Connection reset by peer)
[17:01:20] --> tvm has joined #gemrb
[17:02:23] <tvm> hi, i'm trying to run BG1 under gemrb, it fails with: [KEYImporter]: Opening /mnt/chitin.key...[ERROR], what i'm doing wrong ?
[17:02:41] <pupnik_> hi
[17:02:56] <pupnik_> i think it's not finding your install
[17:02:57] <fuzzie> sounds like your game path isn't pointing at a bg1 install?
[17:03:31] <tvm> well, the problem is that it is not installed, i just pointed it to the CD, guess that's wrong.
[17:04:23] <fuzzie> yes. you can use the original installer (e.g. under wine), or else the wiki has links to install scripts and manual instructions on how to use unshield
[17:04:48] <tvm> ah, right. thanks.
[17:05:06] <tvm> it is probably somewhere in the documentation, but i figured that asking here will be faster.
[17:05:13] * tvm install wine
[17:05:57] <fuzzie> also oh it's a dhewg :P
[17:06:32] <pupnik_> dhewg?
[17:06:57] <dhewg> yes?
[17:07:01] <dhewg> fuzzie: hi :)
[17:09:08] <fuzzie> the errors look like a misinstalled gemrb or maybe a config with GemRBPath or OverridePath pointing to a bad override
[17:10:28] <fuzzie> but lemme go make sure it actually worksd
[17:13:22] <tvm> whee. works.
[17:13:36] <tvm> is there a key to leave fulscreen ?
[17:13:45] <fuzzie> not alt-enter?
[17:14:03] <tvm> nope i guess i tried that.
[17:14:05] <dhewg> ctrl-f
[17:14:05] * tvm tries again
[17:14:06] <fuzzie> ctrl-f, says the source code
[17:14:09] <fuzzie> that's weird :P
[17:14:19] <tvm> yay, ctrl-f works
[17:14:20] <tvm> :>
[17:14:27] <dhewg> boy, its been ages since i played these games
[17:14:57] <tvm> i only know that i played all of them, but barely remember the storyline, etc.
[17:15:03] <fuzzie> i would however in general advise not running master atm
[17:15:15] <tvm> now it's time to play again, after all those years.
[17:15:20] <dhewg> yeah, i compiled from the 0.6.4 branch
[17:15:34] <dhewg> im not sure if the most obvious glitches are known?
[17:15:45] <dhewg> as in: worth reporting?
[17:15:53] <fuzzie> well you are welcome to complain on irc
[17:15:59] <dhewg> hehe
[17:16:22] <dhewg> well, for starters: the gui gfx elements flicker like crazy during battle sometimes
[17:16:25] <fuzzie> there's a lot we know about, but sometimes we miss obvious 'not working' stuff, and bg1/bg2 should be completeable
[17:16:37] <dhewg> one can see the loading gfx at the bottom
[17:16:44] <fuzzie> ah yes, the 0.6.4 branch screws that up
[17:16:53] <dhewg> fixed on master?
[17:17:08] <fuzzie> ab842112 might or might not fix it
[17:17:29] <dhewg> is cherry-picking that safe?
[17:17:38] <fuzzie> althoguh actually it looks insane
[17:17:45] <dhewg> heh :P
[17:18:04] <fuzzie> but yes, it should be safe to cherry-pick and i would love to know if fixes it
[17:18:12] <dhewg> k, i'll try
[17:18:26] <dhewg> next: bg2, that locked portal map with the genie
[17:18:31] <dhewg> kinda at the beginning
[17:18:35] <dhewg> remember that?
[17:18:39] <fuzzie> yeah
[17:18:50] --> boriskr has joined #gemrb
[17:18:53] <dhewg> are all monsters supposed to respawn when reentering?
[17:18:55] <fuzzie> i wasted way too much time going through that stupid dungeon in gemrb :P
[17:19:04] <dhewg> its been so long, i cant remember if it was like that
[17:19:12] <fuzzie> oh ugh, is that broken in 0.6.4?
[17:19:17] <fuzzie> someoen else was reporting that
[17:19:18] <dhewg> well yeah
[17:19:22] <fuzzie> it makes no sense
[17:19:25] <dhewg> same with the droid room
[17:19:57] <tvm> hmm. no sound.
[17:20:02] <dhewg> but worse is, that i entered once, talked to the genie, left, got the bottle, reentered and now i cant spawn him
[17:20:09] <dhewg> so i cant use the bottle it seems
[17:20:26] <fuzzie> all i can theorise is that the map isn't being saved
[17:20:44] <dhewg> i get the circular cursor at that spot, but nothing is happening
[17:21:00] <fuzzie> but it should die in a fit of errors if it can't write to the Cache
[17:21:12] <dhewg> well, the map is completely in fog of war again, is that supposed to be saved?
[17:21:20] <fuzzie> yep
[17:21:27] <dhewg> well thats broken then :P
[17:21:41] <-- pupnik_ has left IRC (Ping timeout: 240 seconds)
[17:22:00] --> pupnik has joined #gemrb
[17:22:00] <-- pupnik has left IRC (Changing host)
[17:22:00] --> pupnik has joined #gemrb
[17:22:23] <boriskr> hi ! I had that problem last week, but when pulled new sources and recompiled gemrb, this problem has gone
[17:22:27] <dhewg> the dialogs on a map restart too
[17:22:34] <dhewg> i guess thats related?
[17:22:41] <dhewg> to the nonsaving part
[17:22:48] <fuzzie> well, who knows :-/
[17:23:05] <dhewg> well im new here, and you're not :P
[17:27:54] <dhewg> did i mention that the most annoying thing is that the option screen settings are not saved?
[17:28:01] <tvm> okay, everything works except the sound.
[17:31:49] <fuzzie> someone should maybe fix the StupidityDetector to try actually writing a file
[17:31:49] <fuzzie> also ok, someone did break bg1 chargen in master
[17:31:49] <fuzzie> great
[17:31:49] <fuzzie> file changes, maybe
[17:32:14] <fuzzie> that is just something we didn't do yet :P
[17:32:49] <fuzzie> ok, tomprince broke 2DA files, i guess
[17:33:46] <fuzzie> "FileStream: Move handling of windows here from VFS" deleted the ReadLine code...
[17:34:22] <fuzzie> and here i was wondering if i am too grumpy about refactoring sometimes :-P
[17:35:45] <fuzzie> that is however a master-only bug.
[17:38:50] <dhewg> is that the bg1 segfault i get?
[17:40:15] <fuzzie> maybe, i'm still trying to work out what's going on for me
[17:40:38] <dhewg> fuzzie: when cherry-picking that gui commit (and fixing the conflict) i get the save screen as savegame thumbnail :P
[17:40:57] <fuzzie> heh, i guess it needs something else
[17:41:19] <fuzzie> i think the problem is that there's some new 'SlicedStream' thing introduced in recent refactoring
[17:41:52] <fuzzie> and it does the wrong thing for 2DA files
[17:43:56] <fuzzie> because it just calls the superclass Read() without paying attention
[17:44:01] <fuzzie> angry fuzzie crush!
[17:44:43] <tvm> both, openal and sdlaudio fail to produce any sound output, no errors though during gemrb start :/
[17:45:51] <tvm> volume is also set correctly, as far as alsamixe allows
[17:45:56] <fuzzie> do you have something like pulseaudio running?
[17:46:42] <tvm> nope, this is very minimalist X11 setup with lightweight wm.
[17:47:14] <tvm> sound works normally in other application
[17:47:18] <tvm> +s
[17:49:03] <fuzzie> well i am a bit distracted fixing that
[17:49:06] <fuzzie> fixing this
[17:49:12] <fuzzie> but i don't think anyone else had that problem, it sounds odd
[17:49:26] <fuzzie> aha, got it
[17:50:09] <tvm> hmm.
[17:50:14] <dhewg> this looks fishy too:
[17:50:15] <dhewg> [ResourceManager]: Searching for *... Tried *.ogg *.acm *.wav *.wav [ERROR]
[17:50:16] <dhewg> [ResourceManager]: Searching for *a... Tried *a.ogg *a.acm *a.wav *a.wav [ERROR]
[17:50:16] <dhewg> [ResourceManager]: Searching for *b... Tried *b.ogg *b.acm *b.wav *b.wav [ERROR]
[17:50:26] <fuzzie> it sounds like you're running master, dhewg
[17:50:41] <dhewg> no its 0.8.4
[17:50:46] <dhewg> well, the branch, not the tag
[17:51:34] <CIA-52> GemRB: 03fuzzie * r5ae9558d3685 10gemrb/gemrb/core/System/SlicedStream.cpp: fix SlicedStream so it doesn't ignore Encrypted setting
[17:51:51] <fuzzie> ^- that fixes the bg1 errors and probably many other errors, many thanks for reporting
[17:52:31] <dhewg> np
[17:52:43] <dhewg> i can play on master to report stuff
[17:52:46] <fuzzie> i think the * stuff is in some of the sound files
[17:52:58] <dhewg> or will that epicbreak my savegames?
[17:53:12] <fuzzie> well, i broke master with some experimental commits to action code, so i would advise against
[17:53:19] <fuzzie> although it does make attacking suck less :P
[17:53:40] <dhewg> fix it and i'll use it? :P
[17:53:57] <fuzzie> yes, the ~2k diff between master and my local repo says hi :p
[17:54:05] <fuzzie> but i got a bit distracted with LB, and real life/work stuff
[17:54:06] <dhewg> :)
[17:54:34] <dhewg> bg2, when killing the 2 golems, i see this repeating with no end in sight: [GameScript]: Aborted action due to death
[17:54:49] <dhewg> 2 timer per sec
[17:55:02] <dhewg> no, one per sec
[17:55:36] <fuzzie> most likely something else i only have fixed locally, unfortunately
[17:58:37] <fuzzie> they'
[17:59:27] <fuzzie> they'll be running igolfle.bcs's MoveToObject block repeatedly due to our buggy death code i guess
[17:59:47] <dhewg> is that bad for my savegame?
[17:59:50] <fuzzie> nope
[18:00:11] <dhewg> k, because when i save at that spot and reload it starts over again
[18:00:21] <fuzzie> oh, that is not so good
[18:00:30] * dhewg reloads
[18:01:10] <dhewg> is there scaling code in gemrb?
[18:01:17] <dhewg> like a dumb 2x?
[18:01:21] <fuzzie> nope
[18:01:26] <dhewg> aww
[18:01:42] <dhewg> i hate fullscreen and 640x480 on full hd is kinda tiny
[18:01:53] <fuzzie> you can apply something like the Widescreen Mod to resize of course, but it stays tiny
[18:02:22] <fuzzie> i imagine you'd be most welcome to patch something up in SDLVideo if looking at the code doesn't cause you to tear your eyes out, of course
[18:03:05] <dhewg> heh, maybe i'll take a look
[18:03:20] <dhewg> its nice to be on the user side for once :P
[18:04:09] <dhewg> but not applying nor saving settings annoys me the most, where do i have to poke at?
[18:04:13] <fuzzie> but gemrb is unfortunately still in the much-loved RE stage of development where attempts to make code nice results only in angry fuzzie complaining about induced bugs for weeks on end
[18:04:52] <fuzzie> we don't have any code at all for saving settings right now, you'd have to work out where to store them etc, that sounds annoying
[18:05:13] <dhewg> how about the used .cfg?
[18:05:23] <dhewg> those settings are supposed to be global, right?
[18:05:45] <fuzzie> yup
[18:06:26] <fuzzie> but you can't always write to that one, sometimes root user install via a package or etc. ~/.gemrb/ would make sense.
[18:06:51] <fuzzie> the lack of settings applying is weirder.
[18:07:32] <fuzzie> it looks like the GUI is just hilariously using the wrong option names
[18:07:34] <dhewg> yeah
[18:07:59] <dhewg> the setting you see on startup do not reflect the settings in use at that moment
[18:09:08] <dhewg> can i do anything about this "genie" situation?
[18:09:35] <fuzzie> could you make sure you actually get the areas written to your Cache directory at runtime?
[18:10:02] <fuzzie> and that you don't get two copies written due to some case-sensitivity issue
[18:10:29] <fuzzie> the genie area is ar0601 i think
[18:10:34] <dhewg> uhm
[18:10:48] <dhewg> that cache, it doesnt need to be persistant, right?
[18:10:53] <fuzzie> nope
[18:10:54] <dhewg> i set it to /tmp/gemrb :P
[18:10:59] <fuzzie> that's fine
[18:11:11] <dhewg> maybe there's a conflict because i switched between master and 0.8.4 a few times?
[18:11:25] <dhewg> the saves are all made with the latter though
[18:11:34] <fuzzie> in fact gemrb wipes the entire directory at load and quit
[18:12:02] <dhewg> i see 69MB after exit :P
[18:12:18] <dhewg> clicking the X though
[18:13:26] <fuzzie> well, we do it in the Interface destructor, so in theory that should delete files too
[18:13:43] <fuzzie> unless your Cache dir is unwritable, hence my questions
[18:13:46] --- Maighstir is now known as Maighstir|away
[18:14:27] <dhewg> its filled fine after rm and restart
[18:14:34] <dhewg> -rw-r--r-- 1 andre andre 70762 2011-04-10 20:12 ar0602.are
[18:14:34] <dhewg> -rw-r--r-- 1 andre andre 70762 2011-04-10 20:12 AR0602.are
[18:14:43] <fuzzie> ok, so that sucks
[18:14:47] <dhewg> :)
[18:14:50] <fuzzie> we already figured something like that was going on
[18:15:05] <dhewg> i set the casesensitivity in the ini though
[18:15:07] <fuzzie> because it crashed when someone accidentally stopped it overwriting files
[18:15:30] --> _pickle has joined #gemrb
[18:15:59] <tomprince> add a strlwn in OpenNew?
[18:17:06] <fuzzie> is that not called anywhere we might care about case-sensitivity?
[18:18:06] <tomprince> that won't work, because we care about paths. But perhaps in the second overload of FileStream::Create
[18:18:16] <fuzzie> i mean, we don't want to force lowercase, either
[18:18:53] <fuzzie> oh, maybe it doesn't matter. i was thinking of saves, but we probably delete/recreate.
[18:19:16] <dhewg> fwiw, wiping the cache didnt help with the genie issue
[18:20:06] <fuzzie> tomprince has the right idea here, on master
[18:20:53] <tomprince> dhewg: What you could do to test is go to the area, delete the file from cache, and then leave. I *think* that would make it work.
[18:21:42] <dhewg> sure
[18:21:44] <dhewg> but uhm
[18:21:46] <-- boriskr has left IRC (Ping timeout: 246 seconds)
[18:21:47] --> boriskr_ has joined #gemrb
[18:21:47] <dhewg> im not sure i understand the real problem
[18:21:51] <dhewg> i mean
[18:21:52] <dhewg> e4233519cb5194c94c4788345331db82 ar0602.are
[18:21:52] <dhewg> 6258bc8307eef26659b0650dd77e6635 AR0602.are
[18:22:09] <fuzzie> one of those areas is the unmodified original one
[18:22:16] <fuzzie> and one of those areas is the one from your running game
[18:22:16] <dhewg> is that genie issue a known bug, and file fixes it but it used the broken file?
[18:22:32] <fuzzie> the bug is that the unmodified original one is there
[18:22:33] <dhewg> *one file
[18:22:36] <fuzzie> ar0602 isn't the genie area though :)
[18:23:01] <dhewg> there's no 601 case conflict though
[18:23:30] <fuzzie> so that is just weird
[18:24:06] <dhewg> there's one .are and one .mos for 601
[18:24:16] <fuzzie> well, assuming you entered the genie area while playing
[18:24:32] <dhewg> i did earlier, without the bottle
[18:24:40] <dhewg> and i could spawn him just fine
[18:24:42] <fuzzie> in the current session, i mean
[18:24:47] <dhewg> yeah, thats borked
[18:25:09] <fuzzie> tomprince: i think you have the right idea there, i am just having difficulty working out exactly what is safe when calling what
[18:25:27] <dhewg> also annoying: when saving in that area after wiping those annoying mosters, there're all back again when loading that again
[18:25:43] <fuzzie> i see that BIFImporter is calling the single-parameter Create with a raw path, and that should be lowercase
[18:26:32] <dhewg> hm
[18:26:43] <fuzzie> and that the other users of CachedCompressedStream are doing strlwr() first
[18:26:48] <dhewg> ok, so i was in the area, wiped the cache, left and reentered
[18:26:55] <dhewg> now i see AR0601.are in the cache dir
[18:27:01] <dhewg> while it was lower case before
[18:27:34] <fuzzie> lowercase comes from a savegame, uppercase probably comes from game data
[18:27:34] <tomprince> remove it, leave and go back, and it should work.
[18:27:50] <dhewg> trying ;)
[18:27:55] * dhewg shakes fist at mephits
[18:30:09] <dhewg> wee, works
[18:30:10] <dhewg> thx :)
[18:31:33] <tomprince> https://gist.github.com/912596
[18:31:40] <dhewg> sword of chaos \o/
[18:32:10] <fuzzie> tomprince: not so useful on 0.6.4
[18:32:13] <dhewg> oh btw, while i see that capital gfx "S" there at the start of the paragraph
[18:32:27] <dhewg> the bg1 intro, where the text scrolls up it broken
[18:32:30] <fuzzie> but also, CacheCompressedStream can't be the problem, right?
[18:32:47] <dhewg> it switches back and forth for the font and gfx letters
[18:37:36] <fuzzie> i don't see how it can be anything but SwapoutArea. but that's weird. no time now.
[18:39:17] <-- boriskr_ has left IRC (Remote host closed the connection)
[18:41:43] <tomprince> The only thing
[18:50:57] <fuzzie> i should say, it's not that the commit above doesn't look ok, although i would want to take a careful look through anything else which might write stuff to CachePath
[18:51:04] <fuzzie> just really should get other stuff done right now
[19:20:37] --> lynxlynxlynx has joined #gemrb
[19:20:38] <-- lynxlynxlynx has left IRC (Changing host)
[19:20:38] --> lynxlynxlynx has joined #gemrb
[19:20:38] --- ChanServ gives channel operator status to lynxlynxlynx
[19:46:30] <dhewg> crasher: char has a full inventory inkl one non-full stacked item (like 30 bolts), let another char pass him that stacked item, which in sums exceeds the max of 40 (like 20 bolts): http://pastie.org/1779993
[19:47:28] <fuzzie> huh, slot -3?
[19:48:15] <fuzzie> which game?
[19:48:18] <dhewg> bg2
[19:48:23] <dhewg> probably related to a similar situation: let that char pick up bolts
[19:48:40] <dhewg> the container then has 40 bolts
[19:49:10] <fuzzie> ok, i guess the -3 is deliberate
[19:49:42] <fuzzie> that's a dumb bug
[19:51:31] <fuzzie> GUIScript.cpp:7387 should have a (Slot >= 0) check
[19:52:13] <dhewg> how about auto-stacks?
[19:52:26] <dhewg> when you pick up or pass stuff
[19:54:16] <fuzzie> that doesn't work?
[19:54:48] <dhewg> i had a single potion, picked up 3 of the same kind at got 2 slots
[19:54:50] <fuzzie> i guess if you have an empty slot first, then it won't fill it
[19:55:02] <fuzzie> should fix that
[19:55:03] <dhewg> and could stack them manually just fine
[19:55:18] <fuzzie> but does it fail if you don't have an empty slot before the existing potion?
[19:55:21] <dhewg> yeah, there was still space in the inventory
[19:55:48] <dhewg> i think that works, except when it crashes :P
[19:55:51] <fuzzie> Inventory::AddSlotItem just goes through every single slot and tries filling them in order, unfortunately
[19:56:35] <fuzzie> could you try adding the (Slot >= 0) check on GUIScript.cpp:7387? the '} else {', if it doesn't match exactly
[19:57:38] <lynxlynxlynx> you're pretty successful at crashing gemrb
[19:57:53] <lynxlynxlynx> (just read the backlog)
[19:57:59] <fuzzie> i'm just impressed at such a comprehensive reproduction recipe :P
[19:58:53] <dhewg> hehe, well i need a few tries to get the recipe down :)
[19:58:53] <fuzzie> dhewg is the magician who has been fixing scummvm's Android support recently, maybe shouldn't be surprised :)
[19:59:04] <dhewg> but i know how it is, coding myself
[20:01:27] <dhewg> another thing: sometimes cutscenes just "scroll away", and you cant stop it
[20:01:41] <dhewg> i guess its related to the cursor pos when it starts
[20:02:27] <fuzzie> should be easy enough to ignore scrolling in cutscene mode :)
[20:02:54] <fuzzie> i forget if you should be able to scroll in dialog mode
[20:05:33] <dhewg> yet another thing: non-dialog text you get on the "?" cursors seems to fade away with a fixed time
[20:05:53] <dhewg> sometimes its lots of text and one cannot read it in time without pause :)
[20:13:02] --- Maighstir|away is now known as Maighstir
[20:14:25] <lynxlynxlynx> was it different in the original?
[20:14:43] <lynxlynxlynx> it also had a subtitle option which helped
[20:14:55] <lynxlynxlynx> i'm not sure we do that yet
[20:17:48] <fuzzie> so far, everything i looked at in the original worked on the basis of the 30fps frames
[20:18:06] <fuzzie> and if you pause, they stop incrementing, so probably fades stop too
[20:20:48] <dhewg> i really cant remember what its like on the original
[21:11:03] <dhewg> fuzzie: bg1 crash is gone on master :)
[21:13:16] <fuzzie> we should put one of those files in the test dataset, i guess
[21:17:03] <dhewg> do you guys work on a stable branch or will master be the next ver?
[21:17:13] <fuzzie> master will be the next ver
[21:19:18] <fuzzie> the dev (Avenger) who does most of the RE code work uses msvc6 on Windows and copies files back/forth to a Linux machine to commit. so that's what the development model revolves around.
[21:20:25] <dhewg> ugh
[21:23:53] <fuzzie> while a nice sane dev model and compiler is nice, in the end, he writes most of the code, so messing with it too much just results in our progress rate dropping through the floor
[21:24:03] <dhewg> changing GUIScript:7386 to "} else if (Slot >= 0) {" fixes the inventory crash for me
[21:24:32] <fuzzie> do the stacks end up correctly distributed?
[21:27:22] <dhewg> nope :P
[21:28:50] <fuzzie> lynxlynxlynx: is that whole GUIScript block just an attempt to hide the bug?
[21:29:48] <lynxlynxlynx> sorry, too tired to look
[21:29:57] <fuzzie> ok, np
[21:30:01] <lynxlynxlynx> up since 0500
[21:30:17] <lynxlynxlynx> lotr scenery though :)
[21:30:20] <fuzzie> hehe
[21:30:31] <-- kettuz has left IRC (Quit: Leaving)
[21:30:40] <fuzzie> i had an early/busy day but no lotr scenery :)
[21:35:13] <fuzzie> but let me try just removing that code entirely and fixing it at source
[21:45:04] <CIA-52> GemRB: 03fuzzie * r5c1fcbf2773e 10gemrb/gemrb/ (core/Inventory.cpp plugins/GUIScript/GUIScript.cpp):
[21:45:04] <CIA-52> GemRB: fix AddSlotItem to calculate partial stack amounts correctly
[21:45:04] <CIA-52> GemRB: this removes the code in GUIScript which looks like it was meant to try
[21:45:04] <CIA-52> GemRB: to fix this, but caused a crash when Slot<0 (reported by dhewg)
[21:45:31] <fuzzie> dhewg: would appreciate you trying that if you have any time, should cherry-pick fine
[21:46:42] <dhewg> sure, sec
[21:48:43] <dhewg> looks good
[21:49:11] <fuzzie> the GUI code probably shouldn't show that string for a partial drop
[21:49:30] <fuzzie> but it's checking IsDraggingItem() rather than the return code of IsDraggingItem() so that's a thing for another day
[21:50:11] <dhewg> it reads "inv full", but it correctly passes the amount to the max stack
[21:50:56] <fuzzie> yay
[21:51:09] <dhewg> and without crashes :)
[21:51:12] <fuzzie> i have no idea how this code works
[21:52:36] <CIA-52> GemRB: 03fuzzie * rb2b717fbd672 10gemrb/gemrb/core/Inventory.h: fix two comments in Inventory.h
[21:52:51] <fuzzie> but hopefull that might help the next person to look at it.
[21:53:46] <dhewg> i just noticed that text on an items description is cut off
[21:54:03] <dhewg> maybe only when the scrollbar is visible
[21:54:06] <fuzzie> hiding under a scrollbar?
[21:54:20] <fuzzie> yes, i think we're probably meant to account for that and hilariosly don't
[21:54:21] <dhewg> and it looks like the width of the bar is just ignored
[21:54:46] <fuzzie> oh dear lord my local code is broken
[21:54:54] <fuzzie> everyone appears to re-die every second
[21:55:06] <fuzzie> i get 3fps
[21:57:15] <dhewg> god, while i remember every detail of adventure games i played on c64/amiga, i dont recall so many bg2 things
[21:57:21] <dhewg> its like a new game \o/
[21:58:03] <lynxlynxlynx> awesome
[21:59:33] <fuzzie> stacks in stores are completely broken
[22:00:06] <fuzzie> my sold stuff just vanishes, the stacks in the store don't exist and my stacks are suddenly worth 0gp once i broke it
[22:00:26] <fuzzie> which is nice because i was wondering how the stack code for stores could possibly work :P
[22:07:09] <fuzzie> i have a fix, will try it in morning
[22:27:19] <-- lynxlynxlynx has left IRC (Remote host closed the connection)