#gemrb@irc.freenode.net logs for 18 Sep 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:29:26] <-- barra_library has left IRC ("Verlassend")
[00:39:57] --> tombhad-AC has joined #gemrb
[00:50:36] <-- tombhadAC has left IRC (Read error: 113 (No route to host))
[01:00:00] <-- tombhad-AC has left IRC ("Verlassend")
[02:48:03] --> Gekz has joined #GemRB
[04:36:07] <-- Maighstir has left IRC (Read error: 131 (Connection reset by peer))
[06:18:15] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[06:40:28] --> Gekz has joined #GemRB
[08:01:46] --> lynxlynxlynx has joined #gemrb
[08:01:47] --- ChanServ gives channel operator status to lynxlynxlynx
[08:06:15] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[08:09:08] --> Gekz has joined #GemRB
[08:50:58] <-- |Cable| has left IRC (Remote closed the connection)
[09:45:11] --> Gekz_ has joined #GemRB
[09:52:32] --> Gekz` has joined #GemRB
[09:54:11] --> barra_library has joined #gemrb
[09:56:52] --> Maighstir has joined #gemrb
[10:01:49] <Maighstir> Could someone take a bug report? Getting errors when trying to compile on Windows
[10:03:05] <Maighstir> K:\__gemrb\gemrb\plugins\Core\ActorBlock.cpp:89: warning: converting of negative value `-0x000000001' to `ieDword'
[10:03:09] <-- Gekz has left IRC (Connection timed out)
[10:03:09] <Maighstir> same on line 91
[10:04:30] <lynxlynxlynx> what are you compiling it with?
[10:04:37] <Maighstir> mingw32
[10:04:57] <Maighstir> as per the insructions on the website... worked a couple of days ago
[10:05:07] <lynxlynxlynx> yeah, this is from a few hours ago
[10:06:45] <Maighstir> Ah, I'm no good with C++, so I wouldn't know how to fix it
[10:06:47] <lynxlynxlynx> change the two lines to use 0
[10:07:11] <fuzzie> not 0xffffffff?
[10:07:13] <Maighstir> right
[10:07:40] <lynxlynxlynx> 0xffffffff is what was expected, but from a quick inspection of the commit it doesn't seem to matter much
[10:11:54] <Maighstir> and then the same thing in plugins\Core\GSUtils.cpp; lines 190 and 195, in case you wish to fix it
[10:20:11] --- Gekz` is now known as Gekz
[10:51:33] <CIA-22> gemrb: 03lynxlupodian * r7195 10/gemrb/trunk/gemrb/GUIScripts/ (bg1/GUIREC.py iwd/GUIREC.py):
[10:51:33] <CIA-22> gemrb: more bg1/iwd guirec syncing: iwd got the second third of customisation, bg1
[10:51:33] <CIA-22> gemrb: a thac0 display fix; -w advised
[11:16:18] <-- Maighstir has left #gemrb ()
[11:16:43] <-- Gekz_ has left IRC (Read error: 104 (Connection reset by peer))
[12:04:43] <CIA-22> gemrb: 03lynxlupodian * r7196 10/gemrb/trunk/gemrb/GUIScripts/ (bg1/GUIREC.py iwd/GUIREC.py): some more guirec merging
[12:07:50] <CIA-22> gemrb: 03lynxlupodian * r7197 10/gemrb/trunk/gemrb/plugins/Core/ (ActorBlock.cpp GSUtils.cpp): fixed compilation on windows
[13:05:35] --> Gekz_ has joined #GemRB
[13:18:40] <-- Gekz has left IRC (Success)
[13:52:55] <lynxlynxlynx> iwd is for lefthanded people
[13:53:09] <lynxlynxlynx> many scrollbars are on the left side of controls
[13:53:49] --> Avenger has joined #gemrb
[13:53:58] <Avenger> hi everyone
[13:54:08] --- ChanServ gives channel operator status to Avenger
[13:56:59] <CIA-22> gemrb: 03avenger_teambg * r7198 10/gemrb/trunk/gemrb/override/bg2/denyspl.2da: area based spell disabling resource (spell vs area => strref)
[13:58:34] <CIA-22> gemrb: 03avenger_teambg * r7199 10/chitem/trunk/ (SpellProperties.cpp chitem.clw chitem.rc resource.h): dltcep update
[14:01:42] <lynxlynxlynx> these are dead magic areas?
[14:02:05] <Avenger> pocketplane/final area/demigorgon prison
[14:02:18] <Avenger> they all disable the pocketplane ability in a hardcoded way
[14:02:51] <Avenger> lynx, you did a change which causes a regression (in theory)
[14:03:06] <Avenger> the place where you disabled null targets for effects
[14:03:21] <Avenger> it will break the summoning wand in irenicus chateau
[14:04:07] <Avenger> i knew it is a potential crasher, but the preferred solution is to handle the null actor in the effects
[14:04:49] <Avenger> there are a few effects (some of which may appear to be arbitrary) where the original engine lets you target non-living scriptables
[14:04:53] <Avenger> one is summoning
[14:04:59] <lynxlynxlynx> yeah, i know it isn't perfect, but i was hitting that crash
[14:05:14] <Avenger> well, you surely can find which effect caused it
[14:05:19] <Avenger> fix the effect :)
[14:06:04] <Avenger> the specification is this: if actor is not null, the target is the actor, if the actor is null then the target is either the area, or the scriptable at the target coordinates, whichever makes sense.
[14:06:08] <lynxlynxlynx> it went bad before, i think
[14:06:25] <lynxlynxlynx> i had two crashes, so i may be mixing them up though
[14:06:27] <fuzzie> yes, i thougt the crash was in effectqueue
[14:06:32] <lynxlynxlynx> one started in the projectile payload
[14:06:41] <Avenger> hi fuzzie :)
[14:06:54] <lynxlynxlynx> brb
[14:07:16] <Avenger> i got another nasty crasher, which is more like fuzzie's concern. In EscapeAreaCore
[14:07:29] <Avenger> it isn't your bug, it is just there ;)
[14:08:05] <Avenger> it looks like the action is executed when the actor is destroyed already
[14:08:33] <fuzzie> huh, that should never happen
[14:08:48] <Avenger> well either that, or something wipes the stack
[14:09:03] <Avenger> the 'current actor' was 0xfeeefeee or something like this in windows :)
[14:09:06] <Avenger> reproducibly
[14:09:37] <fuzzie> does sound more like corruption :)
[14:10:18] <Avenger> hmm, ok, i can look at it in valgrind, i got a save which is 1mp away from the crasher
[14:10:35] <Avenger> 1 sec ;)
[14:10:41] <Avenger> mp is hungarian for sec
[14:12:18] <Avenger> i would reenable the zero actor, those nulls should be handled locally all the way along the execution path
[14:12:20] <lynxlynxlynx> btw, yesterday i encountered an iwd projectile that we don't have yet - it's from color spray, id 149
[14:12:22] <Avenger> ok?
[14:12:39] <Avenger> i implemented some color sprays O_o
[14:12:50] <lynxlynxlynx> iwd uses a different one
[14:12:52] <Avenger> at least 2 different ones
[14:12:56] <Avenger> ok
[14:13:10] <lynxlynxlynx> 149 SPKLARGO
[14:13:18] <lynxlynxlynx> not anywhere else in the tree
[14:13:30] <Avenger> that's not colorspray
[14:13:49] <Avenger> there is an off by one thingie about projectile ids
[14:13:54] <Avenger> sucker bio devs
[14:14:29] <Avenger> so it was the colorspray spell?
[14:14:41] <Avenger> or what thing has the projectile
[14:14:47] <lynxlynxlynx> i think so, it was from a trap
[14:14:49] <Avenger> that's more exact then the number
[14:14:53] <lynxlynxlynx> dltcep showed the same
[14:14:53] <Avenger> hmm
[14:15:03] <lynxlynxlynx> but i have 7.0j
[14:15:10] <Avenger> spklargo is some sparkle area
[14:15:52] <lynxlynxlynx> wouldn't that be spark*
[14:16:55] <lynxlynxlynx> heh, iwd animate dead produces neutral summons
[14:17:17] <fuzzie> Avenger: was the fix to call SetDialog when an area actor had a Dialog ok?
[14:17:27] <Avenger> yes, ithink so
[14:17:29] <fuzzie> i just wonder why it wasn't already there :)
[14:17:35] <Avenger> i don't know why it wasn't there
[14:17:55] <Avenger> if i had time i would check the commit history
[14:18:14] <Avenger> it is pretty trivial to have it, so it must have had some reason
[14:18:19] <fuzzie> i looked through the commit history and it was just missed, as far as i could tell
[14:20:50] <fuzzie> the commit which added the Dialog/Script reading didn't use the variables, you fixed the scripts to be used later, not the dialog.. just as long as there's no reason you remember i think it's fine :)
[14:25:04] <Avenger> lynx can you reproduce the projectile problem?
[14:25:57] <CIA-22> gemrb: 03avenger_teambg * r7200 10/gemrb/trunk/gemrb/override/how/spklargo.pro: missing projectile
[14:26:04] <Avenger> i found a whole bunch of missing projectiles, but they are easy
[14:26:12] <Avenger> if this one works, i copy up all the rest
[14:26:27] <lynxlynxlynx> not easily, i don't have a close save
[14:34:07] <lynxlynxlynx> i saw a proper color spray now too
[14:35:31] <Avenger> this isn't a color spray projectile
[14:35:52] <Avenger> so it is not a big surprise :) Isn't it a bit slow?
[14:36:36] <lynxlynxlynx> i didn't mean the spkabolobla
[14:40:23] <CIA-22> gemrb: 03avenger_teambg * r7201 10/gemrb/trunk/gemrb/plugins/Core/Actions.cpp: fixed a freed pointer reference (never do ReleaseCurrentAction before EscapeAreaCore)
[14:40:41] <Avenger> ok, i fixed the escapearea crasher
[14:41:28] <fuzzie> ah, heh
[14:46:42] <Avenger> and uploading the sparkle area projectiles
[14:47:36] <CIA-22> gemrb: 03avenger_teambg * r7202 10/gemrb/trunk/gemrb/override/ (44 files in 4 dirs): a missing class of standard projectiles
[14:47:51] <Avenger> hmm 44?
[14:48:02] <Avenger> should be 12 per dir...
[14:49:03] <Avenger> ah ok, just 11
[14:52:07] <lynxlynxlynx> they're not in bg2?
[14:55:05] <Avenger> they are
[14:55:19] <Avenger> they are in bg2 in a not hardcoded way
[14:55:36] <lynxlynxlynx> better :)
[14:55:42] <Avenger> actually our .pro and the bg2 .pro differ only in 1-2 bits :)
[14:56:18] <Avenger> these projectiles are very simply to each other, they differ only in the sparkle color
[14:56:32] <Avenger> which is itself not too visible in the original game :)
[15:00:41] <Avenger> heh 22x4 more projectiles are coming
[15:04:53] <Avenger> what is the status of the iwd override ?
[15:04:59] <Avenger> it looks like i didn't add any projectile there
[15:06:01] <lynxlynxlynx> iwd and how are separate
[15:06:42] <Avenger> oh well, this means 354 files duplicated :)
[15:06:58] <Avenger> 354 binary files of size 512-768 :)
[15:07:06] <Avenger> i hope svn won't explode
[15:11:06] <lynxlynxlynx> hit the missing stats from splprot.2da
[15:13:06] <Avenger> sucks
[15:13:19] <Avenger> maybe some of them could be filled in now
[15:13:56] <Avenger> nah, not easily
[15:14:10] <Avenger> 'water using' is a long long list of creature animations :)
[15:14:23] <Avenger> hardcoded
[15:14:28] <Avenger> they so much suck
[15:14:34] <lynxlynxlynx> i don't think it is that one
[15:14:43] <lynxlynxlynx> it comes from the glyph of warding
[15:14:55] <Avenger> evasion check
[15:14:58] <lynxlynxlynx> water/nonwater is something i'd expect with wilting
[15:15:11] <Avenger> evasion check sucks too
[15:15:13] <lynxlynxlynx> yeah, that's only one left
[15:15:24] <Avenger> it is a 'reflex save' squeezed into a 2nd ed game
[15:15:26] <lynxlynxlynx> even the effects the spell uses are unknown
[15:15:34] <Avenger> huh?
[15:15:45] <Avenger> that's most likely configuration problem
[15:15:52] <Avenger> or
[15:16:03] <lynxlynxlynx> oh, no .dat, let me check
[15:16:19] <Avenger> hehe
[15:16:39] <Avenger> skull trap works pretty good already
[15:16:43] <Avenger> glyph is all the same
[15:17:01] <lynxlynxlynx> yeah, that was it
[15:17:10] <Avenger> except this sucky evasion check
[15:17:20] <lynxlynxlynx> huh
[15:17:27] <Avenger> i don't know what it does
[15:17:32] <Avenger> that's why it isn't implemented
[15:17:33] <lynxlynxlynx> sppr304 grants immunity to sppr304
[15:17:51] <lynxlynxlynx> ahh, this is the evasion check :)
[15:18:00] <Avenger> yes, that's standard process to resist stuff
[15:18:21] <Avenger> if the evasion check succeeded, it grants resistance to the rest of the spell
[15:18:30] <lynxlynxlynx> btw, there's also a configuration toggle to have sneak attack instead of backstab ;)
[15:18:59] <CIA-22> gemrb: 03avenger_teambg * r7203 10/gemrb/trunk/gemrb/override/ (442 files in 5 dirs): missing projectiles (iwd had none)
[15:19:03] <Avenger> phew, the massive projectile upload is done
[15:19:17] <lynxlynxlynx> uh
[15:23:18] <lynxlynxlynx> Thief: Thieves improve again, and become better again as a multi-class. At level 7 thieves gain the wonderful "evasion", which enables them to dodge most ... <-- one google hit, but the page and cache are empty, even archive.org doesn't have it
[15:23:37] <lynxlynxlynx> so no idea if it is about icewind dale at all
[15:23:50] <lynxlynxlynx> nor anything concrete
[15:25:53] <lynxlynxlynx> http://www.gamebanshee.com/forums/icewind-dale-11/thief-evasion-skill-102814.html <-- much better
[15:26:37] <lynxlynxlynx> looks pretty much like 3rd edition.
[15:27:36] <lynxlynxlynx> the spell deals damage with two effect calls, could only one be blocked by the immunity?
[15:28:04] <lynxlynxlynx> the spell description says that a save will result in 0 damage
[15:28:16] <fuzzie> "3E Thief Sneak Attack": "enables/disables Sneak Attack and Crippling Stike"? interesting
[15:29:51] <lynxlynxlynx> need to find the manual
[15:30:24] <jcompton> Which manual do you need?
[15:31:45] <lynxlynxlynx> iwd or how
[15:31:58] <lynxlynxlynx> checking the readme now
[15:32:10] <jcompton> Give me a minute, I should have the PDF.
[15:32:43] <Avenger> i should have it somewhere too
[15:32:53] <jcompton> omg race you
[15:34:30] <jcompton> Ah. It turns out that I have the PDF... in Polish only.
[15:34:42] <Avenger> LOL
[15:34:45] <Avenger> how could that be
[15:35:11] <jcompton> I have the CD Projekt DVD compilation, they sent me complimentary copies when they put our mods on their re-releases.
[15:35:29] <jcompton> I had counted on them being lazy and including English docs, but no, they went all-out.
[15:35:46] <Avenger> oh nice
[15:40:36] <jcompton> Found my paper copy, what are you looking for?
[15:40:50] <lynxlynxlynx> thief evasion is hinted in it
[15:41:54] <Avenger> evasion
[15:42:03] <lynxlynxlynx> no need to bother, pdfs are easier to search and i found one on sorcerers
[15:42:04] <Avenger> how is it calculated
[15:42:25] <lynxlynxlynx> Evasion. At 7th level, thieves gain the power to evade effects that would
[15:42:27] <lynxlynxlynx> normally hurt other characters. Whenever a thief is hit by a spell that can
[15:42:28] <lynxlynxlynx> be evaded (Fireball, Lightning Bolt, etc.), the thief is allowed a saving
[15:42:30] <lynxlynxlynx> throw vs. breath to completely avoid the effects of the spell. If the saving
[15:42:32] <lynxlynxlynx> throw is failed, then the spell works normally, including a save for half-
[15:42:33] <lynxlynxlynx> damage if the spell allows for it. Evasion improves as the thief goes up in
[15:42:35] <lynxlynxlynx> level, giving him bonuses to the saving throw vs. breath (for Evasion
[15:42:36] <lynxlynxlynx> purposes only). Keep in mind that thieves cannot evade their own effects,
[15:42:39] <lynxlynxlynx> so mage/thieves should think twice before throwing a fireball at their feet.
[15:42:56] <Avenger> ok, so it is a second save against breath?
[15:44:11] <Avenger> ok, lets see
[15:44:31] <lynxlynxlynx> how it progresses is still unknown
[15:56:01] <Avenger> hmm, do you need to be a level 7 thief, or level 7 and have thief class
[15:56:55] <lynxlynxlynx> i'd say 7 thief levels
[15:57:10] <lynxlynxlynx> so dc/mc chars can have it too
[15:59:13] <Avenger> there is an IE_THIEFLEVEL derived stat, but i don't know if the core works correctly with it
[15:59:48] <lynxlynxlynx> that's a synonym for IE_LEVEL3
[16:00:11] <lynxlynxlynx> use GetThiefLevel()
[16:00:37] <CIA-22> gemrb: 03avenger_teambg * r7204 10/gemrb/trunk/gemrb/ (4 files in 4 dirs): added evasion to spell resistance checks
[16:01:35] <Avenger> oh it is better yeah
[16:03:13] <Avenger> i don't remember doing that one
[16:03:17] <CIA-22> gemrb: 03avenger_teambg * r7205 10/gemrb/trunk/gemrb/plugins/IWDOpcodes/IWDOpc.cpp: Use GetThiefLevel
[16:03:34] <Avenger> most likely because it was made by you?
[16:03:49] <lynxlynxlynx> mattinm
[16:04:26] <Avenger> if it works too, it is excellent :)
[16:04:42] <lynxlynxlynx> it does :P
[16:05:05] <lynxlynxlynx> i'll go try the spell now
[16:05:30] <lynxlynxlynx> got into an avatar removal mess in the meanwhile
[16:06:59] <Avenger> one thing i don't like about it, and other conditional features: this one uses version, and some other use gametype
[16:07:06] <Avenger> it is better to use flags
[16:07:30] <fuzzie> yes, someone should go through and add a lot more flags
[16:07:34] <Avenger> i don't like gametype hardcoded into the engine, it decreases flexibility
[16:08:17] <Avenger> it works well with the existing engines, but if someone, somewhen wants to create a new game, they would better not hack the engine
[16:08:43] <fuzzie> but we got into the habit of abusing existing flags, and that's no good either
[16:09:16] <Avenger> well, depends on the flag, sometimes a flag implies a lot of other changes
[16:09:27] <Avenger> but i don't mind redundancy as long as we have free flags
[16:10:49] <Avenger> this evasion sucks so much. why didn't they add a stat at level 7 (evasion enabled)
[16:10:53] <fuzzie> we could just make it a bool array, it would be just as fast
[16:11:03] <Avenger> i had to hardcode it somewhat
[16:12:04] <Avenger> well, if we are near running out, it should be changed
[16:13:03] <Avenger> feel free to add 20 more flags ;)
[16:14:40] <fuzzie> i can think of a few that i should add, at least
[16:14:57] <fuzzie> and lynx changed some things to check for 'iwd' gamename yesterday that should probably be flags
[16:15:36] <fuzzie> and it looks like i added some code which does that too which needs yet more flags
[16:16:28] <fuzzie> erm, check for 'iwd2', even
[16:16:50] <lynxlynxlynx> mhm
[17:48:52] <Avenger> check this out, it is a good practice to solve spaghetti code: http://www.newgrounds.com/portal/view/511203
[18:10:53] --> |Cable| has joined #gemrb
[18:42:03] <-- Avenger has left IRC ("bye!")
[18:49:44] <CIA-22> gemrb: 03avenger_teambg * r7206 10/gemrb/trunk/gemrb/plugins/Core/Interface.cpp: allow stat symbols instead of stat numbers
[18:52:50] <CIA-22> gemrb: 03avenger_teambg * r7207 10/gemrb/trunk/gemrb/docs/en/Tables/dmgtype.txt: dmgtype documentation
[19:29:12] --> D_T_G has joined #gemrb
[19:31:28] <-- D_T_G has left IRC (Connection reset by peer)
[19:39:27] --> Avenger has joined #gemrb
[19:39:34] --- ChanServ gives channel operator status to Avenger
[19:40:38] <Avenger> it's probably needless to have strings.2da indices in dmgtype.2da, you could use direct strrefs there
[20:23:10] <lynxlynxlynx> if we ever get a shared override folder, then this way is better
[20:24:04] <Avenger> reading the iwd walkthrough bugs
[20:24:31] <Avenger> this should be healed by the dialog fix i guess --> the orcs in the mill before kuldahar (ar2004)
[20:24:47] --> zefklop has joined #GemRb
[20:25:04] <zefklop> hi everyone
[20:25:09] <fuzzie> the kuldahar map bug is a worldmap issue
[20:25:12] <fuzzie> hi zefklop
[20:25:20] <lynxlynxlynx> i'm not sure, i need to check if the area has an override there
[20:26:13] <Avenger> uligar the orc has
[20:26:37] <Avenger> and the cre has no dialog, yeah
[20:26:44] <-- zefklop has left IRC (Remote closed the connection)
[20:27:12] <Avenger> i'm glad i removed a bug from the list without doing work :P
[20:27:29] <fuzzie> the best kind of bug fixes :)
[20:28:04] <Avenger> yeah, close second is when you cut out code and it becomes better :)
[20:28:45] <Avenger> the rest of the bugs don't seem to be trivial, hmm, except the gold thing
[20:29:22] <Avenger> wasn't that a problem in bg2?
[20:29:24] <fuzzie> i guess you can do the gold thing in gui script? just single click -> add gold
[20:29:29] <lynxlynxlynx> many are not iwd specific
[20:30:12] --> zefklop has joined #GemRb
[20:31:30] <Avenger> i don't know how is it done for the containers
[20:34:11] <Avenger> ahh i see, pickupitem is not used for inventory
[20:39:36] <Avenger> the gold thing is fixed :)
[20:42:54] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
[20:46:26] --> Avenger has joined #gemrb
[20:46:31] <lynxlynxlynx> bah, another crash
[20:46:57] <lynxlynxlynx> EffectQueue::ApplyAllEffects calling ApplyEffect with a null *f
[20:47:05] --- ChanServ gives channel operator status to Avenger
[20:47:11] <Avenger> that's odd
[20:47:36] <Avenger> and i didn't even reenable the null target thing yet
[20:47:46] <CIA-22> gemrb: 03avenger_teambg * r7208 10/gemrb/trunk/gemrb/plugins/GUIScript/GUIScript.cpp: donate gold to the party gold instead of dragging it
[20:47:58] <lynxlynxlynx> effects.begin() effects.end()
[20:48:01] <lynxlynxlynx> these are the same
[20:48:16] <fuzzie> that just means it's an empty list
[20:48:21] <lynxlynxlynx> yes
[20:48:23] <fuzzie> any for statement should terminate immediately?
[20:48:31] <Avenger> yes, they should
[20:48:46] <Avenger> where is that
[20:48:57] <lynxlynxlynx> 0xb7fa59e8 in EffectQueue::ApplyAllEffects (this=0x9237e80, target=0x9236e38) at EffectQueue.cpp:478
[20:49:32] <fuzzie> is an effect modifying the list?
[20:49:47] <Avenger> they shouldn't
[20:49:52] <lynxlynxlynx> the backtrace doesn't show any, this was called by RefreshEffects
[20:49:58] <Avenger> but that's the only sane reason
[20:50:09] <fuzzie> it's kind of difficult to debug without knowing which effect was applied the first time through the for loop
[20:50:25] <Avenger> well, the effects should kill themselves or other effects by setting the timing mode to 'just expired'
[20:50:40] <Avenger> the code to remove them is after line 478
[20:51:03] <Avenger> no effect should change the list
[20:51:29] <Avenger> hmm, i mean, remove effects, adding seems to be ok
[20:51:49] <fuzzie> yes, i guess something has to be removing one
[20:52:05] <lynxlynxlynx> i can easily reproduce this
[20:52:15] <Avenger> this is iwd?
[20:52:20] <lynxlynxlynx> i can't put a watch on the thing, since i don't know which actor does it
[20:52:23] <lynxlynxlynx> yes
[20:52:35] <fuzzie> no opcode calls RemoveEffect, which would be my first guess..
[20:52:35] <Avenger> then somehting in iwdopc
[20:52:40] <lynxlynxlynx> i think it is the verbeeg i'm shooting through the wall ;)
[20:52:50] <fuzzie> and other than that and the ApplyAllEffects code, nothing modifies the list?
[20:54:13] <Avenger> AddEffect
[20:54:15] <fuzzie> so unless ApplyAllEffects is being recursively called, which i guess is possible, i have no clue
[20:54:20] <fuzzie> well, i mean, adding things is ok :)
[20:54:43] <Avenger> shouldn't be called recursively either
[20:55:00] <Avenger> so if you find one, that's definitely wrong
[20:55:19] <lynxlynxlynx> hmm, the target is actually an acolyte
[20:55:37] <Avenger> which area?
[20:56:03] <lynxlynxlynx> 3601
[20:56:24] <Avenger> is there combat?
[20:56:30] <Avenger> the acolyte dies?
[20:57:05] <lynxlynxlynx> i was shooting at someone else
[20:57:21] <lynxlynxlynx> i got a segfault now without doing anything after loading
[20:57:27] <Avenger> their only script is TGACOLD
[20:57:33] <Avenger> so it is easy
[20:58:01] <Avenger> huh
[20:58:03] <lynxlynxlynx> now it went a step further to EffectQueue::ApplyEffect (this=0x9211dc0, target=0x9210d78, fx=0x0, first_apply=0) at EffectQueue.cpp:1040
[20:58:06] <Avenger> if you look at the script
[20:58:22] <Avenger> you will see the paramount of blackisle scripting :)
[20:58:32] <Avenger> !Dead("YSELF)")
[20:59:04] <fuzzie> isn't fx_remove_item going to be a potential disaster? or is it okay because it's only called on first apply and not in the loop?
[20:59:32] <lynxlynxlynx> i saw other YSELFs, i thought the decompiler was bugged
[20:59:49] <Avenger> why should it be a problem fuzzie?
[21:00:11] <fuzzie> Avenger: because removing an item can change effects
[21:00:15] <Avenger> no lynx, it is their compiler. they changed the specification of the dead trigger parameter from object to string.
[21:00:20] <fuzzie> when i did that in Die, it recursed and segfaulted
[21:00:26] <fuzzie> maybe i miss something :)
[21:00:26] <Avenger> hmm
[21:00:44] <Avenger> no, it could be a problem
[21:01:04] <fuzzie> we do quite a lot of effects which call inventory which calls RefreshEffect on the actor, though..
[21:01:14] <fuzzie> and it looks like some of those can stay applied, at least
[21:01:27] <Avenger> yes
[21:01:58] <Avenger> maybe time to implement messages
[21:02:55] <Avenger> iwd had this recursive bug too :)
[21:03:02] <Avenger> their animation change reapplies effects, which reapplies the animation change, which loops infinitely
[21:03:17] <Avenger> so i know this hehe :)
[21:04:17] <fuzzie> well, i hope that i'll be better in a few days, and then i could implement some message system, but then it's probably simple enough that you could do it in an hour or so, heh :)
[21:05:28] <Avenger> create inventory item is not bugged, i guess
[21:05:34] <Avenger> only create weapon
[21:06:12] <Avenger> destroyitem is buggy for sure
[21:14:09] <zefklop> hey is anybody working on step sounds?
[21:15:35] <Avenger> not yet
[21:15:55] <Avenger> the sound system leaks, so i didn't make efforts about that
[21:16:03] <zefklop> so consider that now I take this on my todo list :-)
[21:16:07] <Avenger> ok
[21:17:49] <Avenger> see you later
[21:17:50] <-- Avenger has left IRC ("bye!")
[21:32:47] <zefklop> there are leaks in sound manager?
[22:31:51] <-- lynxlynxlynx has left IRC (Remote closed the connection)