#gemrb@irc.freenode.net logs for 21 Jan 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:00:04] <-- edheldil has left IRC (Ping timeout: 248 seconds)
[00:00:20] --> edheldil has joined #gemrb
[00:00:21] --- ChanServ gives channel operator status to edheldil
[00:23:56] <-- edheldil has left IRC (Ping timeout: 240 seconds)
[04:44:49] --> gecko6666 has joined #gemrb
[04:47:26] <-- gecko6666 has left IRC (Client Quit)
[04:47:46] --> gecko6666 has joined #gemrb
[04:47:58] <-- gecko6666 has left IRC (Client Quit)
[04:50:29] --> gecko6666 has joined #gemrb
[04:50:36] <gecko6666> hello
[04:50:56] <gecko6666> i am having a hard time getting gemrb to work on my nexus 7
[04:51:02] <gecko6666> my ersion is
[04:51:15] <gecko6666> can anyone tell me where i can find the log file?
[04:51:23] <gecko6666> i am on android 4.2 btw
[04:57:16] <gecko6666> doesn't seem to be anyone here atm so i'm just gonna go ahead and ask for help here and will check back in later
[04:57:40] <gecko6666> basically i am trying to get either baldur's gate or planescape torment to work
[04:57:47] <gecko6666> both games are the GOG versions
[04:58:00] <gecko6666> i have copied the files over to the nexus
[04:58:22] <gecko6666> but i am guessing the game path in the config file is not correct
[04:58:42] <gecko6666> if i leave it at default, which is "./" it doesn't work
[04:58:52] <gecko6666> nor changing it to the correct path
[04:59:22] <gecko6666> unless it's not the correct path like i think in which case if someone can tell me what it is i would be very grateful
[04:59:50] <gecko6666> btw when i run gemrb it doesn't turn off right away but after perhaps 10 or so seconds
[05:00:06] <gecko6666> anyway will check back in later and thanks in advance for any help
[05:00:23] <gecko6666> i would give log file but i cannot find it anywhere
[05:00:32] <gecko6666> gotta go to bed now
[07:03:35] --> miha has joined #gemrb
[07:14:41] <-- miha has left IRC (Quit: Lost terminal)
[07:20:07] --> edheldil has joined #gemrb
[07:20:07] --- ChanServ gives channel operator status to edheldil
[07:29:40] <-- edheldil has left IRC (Ping timeout: 252 seconds)
[07:29:40] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[08:05:41] --> Drakkar has joined #gemrb
[08:41:17] --> lynxlynxlynx has joined #gemrb
[08:41:17] --- ChanServ gives channel operator status to lynxlynxlynx
[09:33:17] --> edheldil has joined #gemrb
[09:33:19] --- ChanServ gives channel operator status to edheldil
[09:34:47] --> fizzle has joined #gemrb
[10:37:13] <-- WingedHussar has left IRC (Ping timeout: 245 seconds)
[10:38:19] --> WingedHussar has joined #gemrb
[11:45:44] --- ermo is now known as ermo^
[11:54:21] --> rocket_hamster has joined #gemrb
[12:01:20] --> traveler__ has joined #gemrb
[12:03:25] <traveler__> 1. you need alogcat to catch log
[12:03:43] <traveler__> 2. android package is unfortunately very outdated
[12:03:53] <traveler__> 3. path is probably wrong
[12:04:23] <traveler__> 4. while bg1/2 is very much complete, pst is on whole another level of not ready
[12:05:06] <traveler__> 5 not that it doesn't work, it does, but better start with bg1/2
[12:05:30] <traveler__> re: gecko6666
[12:09:37] <Seniorita> [commit] Jens Granseuer: Map::RunAway: fix distance calculation for negative dx/dy http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=88cf913eb34f9370d97f069fcc4a79bc93dcf787
[12:10:54] <lynxlynxlynx> nice
[12:53:37] --> avenger has joined #gemrb
[12:54:17] <avenger> lynx: this one http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commit;h=b026cc8868dbbe009afc61cf04d68366bd3924aa seems to be very unsafe :) Get the string lengths before creating the string
[12:54:18] <Seniorita> SourceForge - gemrb/gemrb/commit
[12:55:14] <lynxlynxlynx> you think 100 is not enough?
[12:56:10] <lynxlynxlynx> anyway, the problem would be clearly visible if some translation managed to bust it
[12:57:13] <avenger> we have good practice examples in the displaymgr, i think
[12:59:14] <lynxlynxlynx> ugly c stuff :P
[12:59:26] <avenger> O_o
[12:59:35] <lynxlynxlynx> strlens
[12:59:44] <lynxlynxlynx> but i guess it could be abstracted
[12:59:53] <avenger> everything goes back to strlen
[13:02:38] <avenger> do iwd2 combat scripts work now, at least somewhat?
[13:02:50] <avenger> i mean mark spell and target
[13:03:50] <avenger> probably not, they miss a big hardcoded part, where the engine decides if a spell is needed to cast :)
[13:05:58] <lynxlynxlynx> pretty well
[13:06:33] <lynxlynxlynx> didn't notice anything out of order
[13:06:59] <lynxlynxlynx> i think visual range will be the next problem
[13:07:29] <lynxlynxlynx> i'm slowly chewing on the inventory front now
[13:08:10] <lynxlynxlynx> action bar part for starters, don't dare touch weapon sets yet
[13:09:02] <lynxlynxlynx> and we have a problem with some of it in other games too, so maybe i should start there
[13:13:43] <Seniorita> [commit] Jaka Kranjc: GemRB_Window_SetupControls: mark empty iwd2 quickitem slots as disabled http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=9fb732862d35331ab2ed85aca7bd76de98cc0f04
[13:13:45] <Seniorita> [commit] Jaka Kranjc: Window_SetupControls: coalesce iwd2 actions later, so the right callbacks get set http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=a7aa3c7a5479b563c8dee772ef36de17043f0a25
[13:14:20] <lynxlynxlynx> oh, one other thing i started contemplating is damage reduction/soaking; it seems all the resistance vs x opcodes work like that in iwd2, not in percentages
[13:14:59] <lynxlynxlynx> opcodes/stats, since some are hardcoded
[13:16:17] <lynxlynxlynx> weapon descriptions still claim to give percentile boni, so that's something to check
[13:23:41] <-- rocket_hamster has left IRC (Remote host closed the connection)
[13:24:09] --> miha has joined #gemrb
[13:37:08] <-- avenger has left IRC (Quit: Page closed)
[13:55:53] <-- Cable_ has left IRC (Ping timeout: 255 seconds)
[14:06:18] <-- traveler__ has left IRC (Ping timeout: 245 seconds)
[14:07:50] --> Cable_ has joined #gemrb
[14:31:16] <-- miha has left IRC (Quit: Lost terminal)
[14:34:08] --> traveler__ has joined #gemrb
[15:57:28] <-- gecko6666 has left IRC (Quit: Page closed)
[16:50:02] <fizzle> lynxlynxlynx: those changes to weapon range you made recently are a little wonky
[16:50:28] <lynxlynxlynx> which part?
[16:50:40] <fizzle> I often have NPCs with ranged weapons stuck on their war cry but never actually attacking anything
[16:51:02] <fizzle> looks like they believe they should be in range but aren't
[16:52:50] <lynxlynxlynx> if you change that 4 to 10, does it fix it?
[16:53:15] <lynxlynxlynx> could be a problem in movenearerto, i didn't check when it is satisfied
[16:54:08] <fizzle> I'll try 10
[17:08:09] <fuzzie> lynxlynxlynx: GetWeapon doesn't set wflags so you can't check that in GetWeaponRange?
[17:08:55] <fuzzie> (since after the more recent commit you no longer call GetCombatDetails in AttackCore)
[17:09:38] <fuzzie> don't know if that's the culprit but it sounds likely, since that means AttackCore gets a *10 multiplier, but PerformAttack gets the proper *4 one since it does call GetCombatDetails
[17:10:17] <lynxlynxlynx> mhm
[17:19:48] --> brada has joined #gemrb
[17:20:39] <lynxlynxlynx> i'll fix it, just first this update thing
[17:40:01] <brada> i dont understand why the chinese bams dont use the first byte for cycle and the second byte for frame...
[17:42:31] <Seniorita> [commit] Jaka Kranjc: iwd2: set ActionsWindow to a fake value, so EF_ACTION works from core too http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=aae29bbd0d8efb8d7c9bd24687b4d12aabc29a2b
[17:42:32] <Seniorita> [commit] Jaka Kranjc: Actor::GetWeapon: set ranged weaponinfo wflag bit, so it can be used in GetWeaponRange http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=5158f4672d412589be70401e6ddba29369683e19
[17:45:56] <fuzzie> brada: well, I imagine they didn't see any reason to do so, when they could just as well use indexes?
[17:48:32] <brada> i told you ttf would be easier :/
[17:48:46] <fuzzie> pretty sure GBK is trivial here
[17:48:52] <fuzzie> take two bytes, convert to index, render, no?
[17:55:07] * fuzzie -> cooking dinner
[18:06:09] <lynxlynxlynx> btw, that other volunteer posted the android build instructions to the tracker
[18:06:18] <Seniorita> [commit] Jaka Kranjc: Actor::IsDualWielding: skip pure ranged weapons http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=eb5686ff2b11e5cd7912b1b082848bd9560ef704
[18:07:09] <brada> lynx: tracker?
[18:07:22] <brada> or just a link? :p
[18:07:44] <lynxlynxlynx> sf tracker :P
[18:07:56] <lynxlynxlynx> https://sourceforge.net/tracker/?func=detail&atid=360122&aid=3599172&group_id=10122
[18:07:58] <Seniorita> SourceForge.net: GemRB Game Engine: Detail: 3599172 - Android Build
[18:08:06] <brada> ah the bug tracker
[18:08:48] <brada> ill look at that after this font business is done
[18:10:05] <lynxlynxlynx> miha volunteered to try this week
[18:10:28] <brada> fuzzie: they do use first byte for cycle and second for frame
[18:10:34] <brada> but you have to do cycle - 1
[18:10:37] <brada> :D
[18:10:40] <lynxlynxlynx> anyway, this is now a release blocker if you ask me (not that one's very near)
[18:10:52] <brada> lynx: i agree
[18:17:12] <-- brada has left IRC (Quit: brada)
[18:22:43] <-- edheldil has left IRC (Remote host closed the connection)
[18:23:13] <fuzzie> brad: hoera!
[18:38:21] <Seniorita> [commit] Jaka Kranjc: Effect: renamed Probability{1,2} for clarity http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=e7eb349a9cbb93f798049ef68648f1f4e0f1f81e
[18:47:39] <Seniorita> [commit] Jaka Kranjc: iwd2::guirec: don't display damage boni with 0 dice values http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=550178e3514e1992be62c736d5bf980775070615
[19:03:48] <-- Cable_ has left IRC (Ping timeout: 248 seconds)
[19:16:53] --> Cable_ has joined #gemrb
[19:23:53] --> brada has joined #gemrb
[19:35:09] <traveler__> wow. those buld instructions for android surely looks convulted.
[19:35:59] <traveler__> *build *convoluted
[19:39:04] --> Yoshimo has joined #gemrb
[19:53:58] <fuzzie> well, the problem is that you have to build the whole unofficial SDL *and* make it use gemrb *and* patch gemrb, plus all the normal complexity of building for android
[19:55:16] <fuzzie> but it's all first-time pain
[19:55:29] <lynxlynxlynx> i doubt you still have to patch gemrb, unless you want a different version string
[19:55:56] <fuzzie> is that using thing applied? it seems weird
[19:56:28] <fuzzie> doesn't look like it, so the patch seems still needed
[19:57:11] <fuzzie> (since 'core' is in GemRB namespace)
[19:58:09] <fizzle> lynxlynxlynx: that range thing might be more serious
[19:58:20] <fizzle> mages seem to be affected, too
[19:58:42] <lynxlynxlynx> you mean spell range?
[19:58:45] <fizzle> half my spells are simply vanishing into thin air
[19:58:51] <fizzle> looks like it, anyway
[19:58:54] <fuzzie> it shouldn't affect spell range
[19:58:57] <lynxlynxlynx> can't be related
[19:59:12] <fizzle> yeah, doesn't look like your patch
[19:59:18] <lynxlynxlynx> the last thing i did about spell ranges was actually increase it
[19:59:35] <lynxlynxlynx> but both have magic constants involved
[19:59:50] <lynxlynxlynx> some kind of wierd pixels to feet or viceversa conversion
[20:00:00] <Seniorita> [commit] Jaka Kranjc: Inventory::GetUsedWeapon: fixed offhand lookup and resimplified IsDualWielding http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=ee2f5d543275f285174899a6ad3440daa747d634
[20:00:15] <fuzzie> lynxlynxlynx: isn't wflags uninitialised? (so ORing will be unsafe)
[20:01:01] <Seniorita> [commit] Jens Granseuer: this one slipped through last time http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=749d6a8cea9d4e7cc517144246cb92df093d35a1
[20:01:02] <Seniorita> [commit] Jens Granseuer: make the confirmation window modal to prevent clicks on parent window http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=b24c69cdef448cf6bfd849d00b8c9cf25ac40cc6
[20:02:46] <lynxlynxlynx> you're right
[20:02:54] <lynxlynxlynx> i should just make it a class
[20:03:45] <lynxlynxlynx> it's not used for other things there though
[20:03:48] <fuzzie> well, you can give a constructor to a struct
[20:04:08] <fuzzie> if you just want to zero fields
[20:04:25] <fuzzie> well you can give a constructor to a struct anyway but I mean it doesn't seem too bad to just zero fields :)
[20:05:49] <lynxlynxlynx> when i started the combatinfo stuff, i thought i'd end up rewriting weaponinfo too, but in the end it didn't look necessary
[20:06:40] <fuzzie> right. you could put a simpler struct inside the more complex struct if you wanted to be 'nice' and not have a bunch of invalid fields, I guess.
[20:07:10] <lynxlynxlynx> just zeroing everything sounds good enough
[20:07:10] <fuzzie> but I'm just trying to avoid bugs here, don't have any issues with the design :)
[20:07:33] <lynxlynxlynx> avenger would say it is the responsibilty of the caller >>
[20:07:57] <lynxlynxlynx> that's why we don't init all the item or item header fields
[20:14:57] --> edheldil has joined #gemrb
[20:14:57] --- ChanServ gives channel operator status to edheldil
[20:19:11] --- ermo^ is now known as ermo
[20:21:34] <brada> https://github.com/bradallred/gemrb/tree/mbfonts
[20:21:36] <Seniorita> bradallred/gemrb at mbfonts · GitHub
[20:21:42] <brada> im ready to merge if that looks good
[20:24:34] <brada> :/
[20:25:10] <brada> somewhere i broke something
[20:25:18] <fuzzie> oh?
[20:25:36] <brada> yeha on the char info screen im getting text that should be parsed out
[20:26:11] <brada> and the status icons are in rows of 2 o_O
[20:26:21] <brada> i dont even have a guess at how that happened
[20:27:20] <lynxlynxlynx> git bisect is the easiest way out, you don't even have to think :)
[20:27:47] <lynxlynxlynx> the summary of this branch would be complete mb support?
[20:28:08] <brada> as complete as i know to make it
[20:29:40] <fuzzie> the bam resref commit seems pretty dangerous
[20:30:12] <brada> i guess so
[20:30:20] <brada> technically
[20:30:21] <fuzzie> in the sense that it'll quietly corrupt things if something opens a bam from non-resref form
[20:31:13] <fuzzie> oh, I guess it also doesn't work
[20:31:28] <fuzzie> since you'll copy STATES.BAM to STATE\0?
[20:31:45] <fuzzie> so that might be your problem!
[20:32:38] <fuzzie> oh, no, it won't, will it.
[20:32:39] <fuzzie> never mind me.
[20:32:49] <fuzzie> strncpy is the one I hate.
[20:34:30] <brada> ok
[20:36:21] <brada> so the problem i just had is easily fixed by not having Encoding=chinese in the cfg when using english tlk
[20:36:44] <brada> but i also had a bug in dbStreLen making it return len+1
[20:38:30] <fuzzie> ah
[20:38:32] <fuzzie> all fixed?
[20:38:39] <brada> seems so
[20:39:10] <fuzzie> i was going to say that the states are looked up by ord(c)-66, and the highest entry in the statdesc file is 145, so that means the chars go up to 211
[20:41:07] <-- edheldil has left IRC (Ping timeout: 248 seconds)
[20:42:22] <fuzzie> so maybe you want to special-case states to not multibyte if you didn't already and I missed it
[20:44:41] <brada> hmm
[20:45:00] <brada> why wont they work as is?
[20:45:24] <brada> oh
[20:45:25] <brada> yeah
[20:45:29] <brada> duh
[20:59:27] <lynxlynxlynx> i don't geet it
[20:59:31] <lynxlynxlynx> cmake 41 minutes ago
[20:59:41] <lynxlynxlynx> yet there's nothing in the commit list view
[21:01:00] <-- Yoshimo has left IRC (Quit: Yoshimo)
[21:01:02] <lynxlynxlynx> and the parent is from several days ago
[21:01:13] <fuzzie> .. ?
[21:02:01] <lynxlynxlynx> github
[21:02:08] <fuzzie> ah
[21:04:05] <fuzzie> brad probably rebased it
[21:04:23] <brada> yeah
[21:07:34] <lynxlynxlynx> tell me once you're done and i'll do a compile test
[21:09:12] <brada> well you can test the compile now
[21:09:28] <brada> im just working out the states font stuff
[21:20:17] <lynxlynxlynx> missing includes
[21:20:35] <lynxlynxlynx> TTFFont.cpp:50:41: error: ‘errno’ was not declared in this scope
[21:21:41] <lynxlynxlynx> ah no, should be ret i guess
[21:23:54] <lynxlynxlynx> uint8_t is also not defined here
[21:24:16] <brada> interesting
[21:28:07] <lynxlynxlynx> and once you fix it, please merge the fixes into commits that introduced the code
[21:37:56] <brada> yeah i will rebase
[21:43:29] --> edheldil has joined #gemrb
[21:43:30] --- ChanServ gives channel operator status to edheldil
[21:43:49] <-- edheldil has left IRC (Client Quit)
[21:43:56] --> edheldil has joined #gemrb
[21:43:57] --- ChanServ gives channel operator status to edheldil
[21:45:19] <Seniorita> [commit] Jaka Kranjc: zero WeaponInfo fields on creation http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=7d9270d1809d4c722b9f09df92844563136af23d
[21:45:20] <Seniorita> [commit] Jaka Kranjc: PCStatsStruct::InitQuickSlot: added debug print in case of unhandled value http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=36a11188d34f7478427009881242928b8bc1d834
[21:59:45] <Seniorita> [commit] Jens Granseuer: also reset target mode on pressing escape http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=645c5ff413ea09a252bdf2aa06a5c58b2b7b4b51
[21:59:46] <Seniorita> [commit] Jens Granseuer: get rid of a few more unnecessary empty buttons http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=bdd7f239c53945a12c92c8e16c574f9bb4313d2f
[22:09:38] <traveler__> hm. tazok can hit for 46hp critical
[22:10:54] <lynxlynxlynx> nice
[22:11:18] <lynxlynxlynx> probably doesn't use a normal weapon
[22:11:47] <lynxlynxlynx> would need to be a 2d12 or better for that effect
[22:13:27] <traveler__> some two handed
[22:13:29] <traveler__> sword
[22:13:34] <traveler__> + being half ogre?
[22:14:40] <traveler__> i just noticed that asleep/being knocked back doesn't change avatar properly
[22:14:44] <lynxlynxlynx> magic 2h, godlike str and other effects could do it too
[22:14:49] <traveler__> you just stand there
[22:15:03] <lynxlynxlynx> but he may just have the magic slot with something ridiculous
[22:15:15] <traveler__> i think there should by lying anim
[22:15:27] <traveler__> i will dump him when he stops being invisible
[22:15:28] <traveler__> now
[22:15:29] <lynxlynxlynx> yeah, it's the same as the dying one
[22:15:37] <traveler__> i think
[22:16:17] <lynxlynxlynx> i doubt we skip pcs, it's more likely something resets the stance
[22:16:21] <traveler__> console spammed with 'bat- unconscious' poor mother nature!
[22:17:57] <traveler__> http://pastebin.com/SkR9rhng tazok
[22:17:59] <Seniorita> [Actor/DEBUG]: Debugdump of Actor Tazok (Tazok, Tazok): Scripts: shout <none> < - Pastebin.com
[22:23:37] <traveler__> why there is no icon for onvisibility on portriat + no info on character sheet?
[22:24:10] <traveler__> not sure about gfx effect as i have this magic globe on
[22:32:34] <traveler__> what is efect of critical mis in ie games?
[22:32:38] <traveler__> *miss
[22:32:53] <traveler__> does antagonist have extra attack?
[22:44:07] <lynxlynxlynx> no
[22:44:20] <lynxlynxlynx> you roll a 1
[22:44:29] <lynxlynxlynx> you can't hit a building
[22:44:38] <lynxlynxlynx> right in front of you
[22:44:50] <lynxlynxlynx> automatic failure
[22:45:17] <lynxlynxlynx> like you can critically hit an uranium core
[22:56:15] <brada> lynx: all rebased
[22:56:16] <brada> https://github.com/bradallred/gemrb/compare/master...bradallred:mbfonts
[22:56:22] <brada> should build for you now
[22:56:24] <Seniorita> Comparing master...mbfonts · bradallred/gemrb · GitHub
[22:57:16] <brada> kind of strange that iconv doesnt include errno itself since thats what it stores its error in...
[22:59:31] <lynxlynxlynx> ok, let's see
[23:00:05] <lynxlynxlynx> crap
[23:00:16] <lynxlynxlynx> forgot to stash my changes
[23:02:08] <lynxlynxlynx> ok, it does compile
[23:02:19] <lynxlynxlynx> i get a crash in iwd2 though
[23:02:23] <lynxlynxlynx> maybe due to states
[23:02:32] <brada> :/
[23:02:35] <lynxlynxlynx> all was fine until the main contents of the records screen
[23:02:38] <lynxlynxlynx> [BAMFont/ERROR]: normal missing glyph for character '7c7c' using ISO-8859-1 encoding.
[23:02:38] <lynxlynxlynx> *** Error in `gemrb/gemrb': malloc(): smallbin double linked list corrupted: 0x000000000121acb0 ***
[23:02:46] <lynxlynxlynx> missing glyph spam
[23:02:57] <brada> thats not for states tho
[23:03:17] <lynxlynxlynx> i don't have the start in my history
[23:03:20] <lynxlynxlynx> it tried too many
[23:03:40] <lynxlynxlynx> you sure you're not doubling the bytes or something? all are in pairs it seems
[23:04:08] <brada> im sure i dont get that on my mac with bg2
[23:04:12] <lynxlynxlynx> the last thing in my buffer is [BAMFont/ERROR]: normal missing glyph for character 'afaf' using ISO-8859-1 encoding.
[23:04:32] <brada> all of the characters are the same byte twice?
[23:04:37] <lynxlynxlynx> yep
[23:04:38] <brada> strange
[23:04:59] <lynxlynxlynx> maybe it's the 2
[23:05:28] <lynxlynxlynx> i saw the old code for states2, but are you now copying it badly now maybe?
[23:05:50] <brada> if you got a malloc error then somewhere somthing like that is happening
[23:06:08] <brada> those errors were for the normal font tho not states
[23:06:13] <lynxlynxlynx> btw, i did saw a + in the portraits
[23:06:25] <lynxlynxlynx> and bg2 crashed on load
[23:06:39] <brada> clearly something behaves diffrently on my machine
[23:06:48] <lynxlynxlynx> [BAMFont/ERROR]: normal missing glyph for character '96e0' using ISO-8859-1 encoding.
[23:06:48] <lynxlynxlynx> [BAMFont/ERROR]: normal missing glyph for character '3d3' using ISO-8859-1 encoding.
[23:06:48] <lynxlynxlynx> [BAMFont/ERROR]: normal missing glyph for character '96e0' using ISO-8859-1 encoding.
[23:06:48] <lynxlynxlynx> [BAMFont/ERROR]: normal missing glyph for character '3d3' using ISO-8859-1 encoding.
[23:06:49] <lynxlynxlynx> *** Error in `gemrb/gemrb': double free or corruption (out): 0x0000000003d70000 ***
[23:07:26] <brada> kinda hard for me to fix since i have no problems
[23:07:27] <lynxlynxlynx> _ZNK5GemRB4Font13PrintFromLineEiNS_6RegionEPKhPNS_7PaletteEhPS0_PNS_8Sprite2DEjb+0xc0b)[0x7f41add38981
[23:07:37] <lynxlynxlynx> i can grind it tommorow
[23:07:42] <brada> cool
[23:07:46] <brada> thanks
[23:08:29] <lynxlynxlynx> eh, i'll just do it now
[23:08:39] <lynxlynxlynx> i forget this machine is not so slow
[23:11:56] <lynxlynxlynx> cool, managed to even get garbage in the journal
[23:13:46] <lynxlynxlynx> all invalid reads it seems
[23:15:06] <-- fizzle has left #gemrb
[23:15:10] <brada> but where?
[23:16:43] <lynxlynxlynx> http://sprunge.us/gAZV
[23:16:53] <lynxlynxlynx> like before, PrintFromLine
[23:17:10] <lynxlynxlynx> i suggest you download that and use less -R
[23:18:48] <-- traveler__ has left IRC (Ping timeout: 245 seconds)
[23:21:31] <brada> something seems wrong
[23:21:41] <brada> line 117 of font.cpp is a }
[23:23:06] <lynxlynxlynx> why look at 117?
[23:23:16] <lynxlynxlynx> 244 is reported
[23:24:47] <brada> all those suggest that len is wrong
[23:24:57] <brada> i cant possibly see how
[23:26:08] <lynxlynxlynx> tomprince is approximately in your timezone and he's good at this too
[23:26:16] <lynxlynxlynx> me, zzz
[23:26:19] <brada> ok
[23:27:36] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:30:21] --> traveler__ has joined #gemrb
[23:39:29] <brada> i cant even run valgrind anymore :/
[23:39:41] <brada> just get a "the 'impossible' happened" error
[23:47:24] <brada> "WARNING: Support on MacOS 10.8 is experimental and mostly broken."
[23:47:29] <brada> well that explains it :/