#gemrb@irc.freenode.net logs for 26 Dec 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:03:54] <-- Avenger has left IRC (Quit: bye!)
[00:29:19] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[01:16:46] <-- kida_laptop has left IRC (Remote host closed the connection)
[02:37:40] <-- CamDawg has left IRC (Quit: Leaving.)
[02:39:57] --> CamDawg has joined #gemrb
[02:41:11] <-- CamDawg has left #gemrb
[03:48:58] <-- brada has left IRC (Quit: brada)
[04:10:11] --> brada has joined #gemrb
[04:17:52] <-- brada has left IRC (Quit: brada)
[04:46:35] --> brada has joined #gemrb
[04:48:08] <-- brada has left IRC (Client Quit)
[04:50:40] --> brada has joined #gemrb
[04:53:06] <-- brada has left IRC (Client Quit)
[05:42:28] <Gekz> I have this sneaking suspicious that the infinity engine is a poorly coded piece of crap
[05:42:37] <Gekz> and the fact it worked at all is amazing
[07:56:18] <edheldil_> no, it's a trap (DC20, Will) to lure in naive reimplementors. Each year, you have to do a check DC25 vs INT or suffer insanity for 1d3+2 months
[08:07:41] <-- |Cable| has left IRC (Ping timeout: 255 seconds)
[08:21:44] --> |Cable| has joined #gemrb
[08:33:14] <-- edheldil_ has left IRC (Ping timeout: 255 seconds)
[09:46:06] --> lynxlynxlynx has joined #gemrb
[09:46:06] --- ChanServ gives channel operator status to lynxlynxlynx
[10:26:18] <lynxlynxlynx> hah, currently some spellcasters are the best fighters in iwd2
[10:27:05] <lynxlynxlynx> two of the five damage type boni are misapplied, so they get 20 or 40 regular damage bonus
[10:32:15] <gembot> build #158 of osx-xcode-binary is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/osx-xcode-binary/builds/158 blamelist: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[10:37:24] <lynxlynxlynx> yep, moved
[11:07:52] --> Den_ has joined #gemrb
[11:08:00] <Den_> Привет
[11:08:07] <Den_> Парни нужна помощь
[11:08:36] --> edheldil_ has joined #gemrb
[11:08:46] <Den_> Кто живой?
[11:12:10] <edheldil_> Den_: me, for example.
[11:12:34] <Den_> HI, Do you speek Russian?
[11:13:04] <edheldil_> very little. But we can try
[11:14:07] <Den_> Ok, i need help, with IceWind Dale for android
[11:14:08] <edheldil_> But I don't have russian keyboard, so you I will reply in English
[11:14:19] <Den_> I try english
[11:14:29] <edheldil_> what problem?
[11:14:49] <Den_> Gem don't work, kick me
[11:14:59] <edheldil_> hmm, what do you mean?
[11:15:27] <Den_> When I enter in Gem, 2 seconds and the end(
[11:16:29] <Den_> Two days i have sex with gem)))
[11:17:27] <lynxlynxlynx> market version or from sourceforge?
[11:17:36] <Den_> market
[11:17:57] <Den_> so i download from playmarket
[11:18:36] <lynxlynxlynx> do you have a GemRB.log where you installed icewind dale
[11:19:22] <Den_> GemRB.cfg???
[11:19:34] <lynxlynxlynx> no
[11:19:35] <lynxlynxlynx> ok
[11:19:49] <lynxlynxlynx> gemrb closes so fast, since it can't find the files
[11:20:03] <lynxlynxlynx> GemRB.cfg
[11:20:20] <lynxlynxlynx> in there, does GamePath show where the game is installed?
[11:20:36] <lynxlynxlynx> is GameType set to "how" or iwd?
[11:20:46] <Den_> How about CD path?
[11:21:00] <Den_> GameType i write
[11:21:01] <edheldil_> Den_: depends on your setup
[11:21:23] <edheldil_> Den_: post your GemRB.cfg to http://pastebin.ca
[11:24:14] <Den_> GamePath=/sdcard/app-data/net.sourceforge.gemrb/ CD1=./
[11:24:37] <Den_> other disk is empty to
[11:24:37] <lynxlynxlynx> not on the same line, right?
[11:24:48] <Den_> no
[11:24:51] <lynxlynxlynx> ok
[11:24:54] <lynxlynxlynx> that's fine
[11:24:56] <Den_> different
[11:25:03] <lynxlynxlynx> did you copy over a full installation?
[11:25:34] <Den_> yes
[11:25:36] <lynxlynxlynx> you know, when you install iwd you have the choice of full or minimal (needs cd)
[11:25:39] <lynxlynxlynx> ok
[11:26:01] <Den_> on PC i don't need CD
[11:26:23] <lynxlynxlynx> besides the config like edheldil_ asked, can you also post the file listing of /sdcard/app-data/net.sourceforge.gemrb/ ?
[11:29:12] --> traveler has joined #gemrb
[11:29:39] <Den_> Yes, where i can post my file?
[11:29:51] <lynxlynxlynx> http://pastebin.ca for example
[11:30:14] <Den_> Ok i try
[11:30:16] --> rocket_hamster has joined #gemrb
[11:32:46] <Den_> http://pastebin.ca/2296770
[11:32:53] <Den_> i did
[11:35:31] <lynxlynxlynx> uncomment CaseSensitive=1
[11:35:48] <Den_> i must enter 1???
[11:36:41] <lynxlynxlynx> remove #
[11:36:58] <Den_> Did it
[11:37:40] <lynxlynxlynx> check if it helps
[11:40:34] <Den_> I try, don't helps
[11:41:49] <Den_> #GameOnCD=0 i don't need CD
[11:42:00] <Den_> Should i rewrite this
[11:43:45] <Den_> Maybe problem is in gamepath???
[11:48:19] <traveler> GamePath=/sdcard/app-data/net.sourceforge.gemrb/
[11:48:22] <traveler> looks normal
[11:48:25] <lynxlynxlynx> yeah
[11:48:35] <lynxlynxlynx> what do you have there?
[11:49:04] <Den_> Where?
[11:49:08] <traveler> GamePath=/sdcard/app-data/net.sourceforge.gemrb/
[11:49:25] <lynxlynxlynx> is Chitin.key in there?
[11:49:34] <Den_> file?
[11:49:45] <traveler> yes
[11:50:05] <Den_> yes it'is
[11:58:52] <Den_> Den waits, he knows, that gods helps him!
[12:04:30] <lynxlynxlynx> @205 = ~Thanks. Vlah blah blah. You are great. Blah blah blah. Money. Experience. Chicks. All yours. Testocles happy now.~ <-- iwd2 test char
[12:04:55] <lynxlynxlynx> Den_: since you don't have a log file, please install alogcat
[12:06:02] <Den_> logcat what is it?
[12:06:27] <Den_> what log file?
[12:09:26] <traveler> alogcat
[12:09:29] --> CamDawg has joined #gemrb
[12:09:37] <traveler> logging utility for android
[12:10:00] <traveler> usually gemrb lefts log
[12:10:26] <traveler> but if you are on old version from market, you need alogcat iirc
[12:11:06] <Den_> where i can dowload this?
[12:11:14] <traveler> i don't know
[12:11:20] <lynxlynxlynx> market
[12:11:25] <traveler> probably from whre you got gemrb
[12:11:29] <traveler> market that is
[12:12:57] <Den_> So reinstall GemRB?
[12:15:55] <edheldil_> no, install alogcat app from the market and run GemRB again
[12:15:58] <traveler> install alogcat
[12:16:33] <traveler> run alogcat instantly after gemrb did not start
[12:18:16] <Den_> I did
[12:18:30] <edheldil_> post the log to pastebin
[12:20:24] <traveler> *going afk for a bit
[12:23:27] <Den_> http://pastebin.ca/2296780
[12:25:14] <lynxlynxlynx> Cache path ./Cache doesn't exist, not a folder or contains alien files
[12:25:18] <lynxlynxlynx> clear it
[12:26:15] <Den_> Where?How?
[12:26:22] <lynxlynxlynx> some file manager
[12:26:37] <lynxlynxlynx> we should just tell Beholder to use a different default cache path
[12:27:08] <Den_> I must create path in GemRB config?
[12:27:21] <lynxlynxlynx> no no no
[12:27:27] <lynxlynxlynx> delete Cache/*
[12:30:17] <Den_> CachePath= like this?
[12:30:45] <lynxlynxlynx> no, do not change the config
[12:31:08] <lynxlynxlynx> remove the contents of the directory
[12:31:10] <traveler> delete all files you have now
[12:31:19] <traveler> in cache directory
[12:31:36] <Den_> that file from game
[12:32:37] <edheldil_> I/printf: ( 5534): **contains another dir**
[12:32:37] <edheldil_> I/GemRB ( 5534): Cache path ./Cache doesn't exist, not a folder or contains alien files!
[12:33:17] <edheldil_> it says that your cache directory contains file(s) that should not be there. GemRB aborts to be safe
[12:36:10] <Den_> Go Go Go all play!!! Thanks very much))))
[12:36:39] <edheldil_> :)
[12:39:29] --> Yoshimo has joined #gemrb
[12:46:10] <lynxlynxlynx> is there a better way to compute a base 2 log than by dividing two logs? Would a loop with shifting be preffered?
[12:47:12] <lynxlynxlynx> i don't need the remainder
[12:49:41] <wjp> yes, a loop with shifting would be preferred
[12:52:37] <lynxlynxlynx> ok
[13:13:35] <-- rocket_hamster has left IRC (Ping timeout: 255 seconds)
[13:31:41] <-- |Cable| has left IRC (Ping timeout: 245 seconds)
[13:40:41] --> rocket_hamster has joined #gemrb
[13:42:50] <-- Yoshimo has left IRC (Quit: Yoshimo)
[13:45:02] --> |Cable| has joined #gemrb
[13:46:36] --> Yoshimo has joined #gemrb
[13:50:03] <rocket_hamster> in python || has higher priority then = ?
[13:52:47] <rocket_hamster> == even
[14:12:25] <edheldil_> in python there's no '||', is thre?
[14:14:23] <-- Yoshimo has left IRC (Quit: Yoshimo)
[14:18:35] <rocket_hamster> well there is 'or'
[14:21:30] --> kida_laptop has joined #gemrb
[14:21:35] <-- Drakkar has left IRC ()
[14:39:23] --> Drakkar has joined #gemrb
[14:55:09] <traveler> what's up with
[14:55:14] <traveler> irc commit bot?
[15:08:17] <lynxlynxlynx> long dead
[15:10:05] <lynxlynxlynx> the easiest solution would be to use the github bot i guess, but that mirror updates rarely - for example now, the last update time was 7h ago
[15:13:59] <lynxlynxlynx> ah no, got a better idea
[15:14:23] <lynxlynxlynx> we could just read the rss from gitweb
[15:14:57] <lynxlynxlynx> i'll ask around, i know a few people with scrappers
[15:24:25] <-- |Cable| has left IRC (Ping timeout: 252 seconds)
[15:25:45] <traveler> it's not that important for me, probably would mitigate compulsively pulling git when on irc though
[15:27:28] <lynxlynxlynx> we miss it, it's good for discussion and spurring review
[15:32:32] <traveler> indeed
[15:37:10] --> |Cable| has joined #gemrb
[15:41:16] --> Avenger has joined #gemrb
[15:41:17] --- ChanServ gives channel operator status to Avenger
[15:41:37] <Avenger> hi
[15:41:55] <traveler> hello
[15:42:55] <Avenger> so lynx you miss irc commits too? i still think our gembot could help somehow
[15:43:21] <Avenger> it sees new commits, and it can drop lines on the channel, so, it seems trivial
[15:43:38] <lynxlynxlynx> but that would require extra work, i think it can be done in a simpler manner
[15:43:56] <Avenger> even simpler?
[15:44:04] <Avenger> if i had its source i could do it :D
[15:44:14] <lynxlynxlynx> buildbot is open source
[15:44:44] <lynxlynxlynx> and it easily reports to irc, but the last time we tried, it reported for each build slave and so was too noisy
[15:45:07] <Avenger> well, pick a build slave
[15:45:14] <Avenger> and filter messages for it
[15:45:29] <Avenger> i know it is lame hack, but it would work
[15:46:04] <Avenger> where is your partially done code?
[15:47:24] <lynxlynxlynx> i didn't code anything, i just ordered it around
[15:47:54] <Avenger> ah, so it can do all the stuff except it cannot watch a single slave?
[15:47:58] <lynxlynxlynx> gembot: notify on started
[15:47:59] <gembot> The following events are being notified: ['started', 'exception', 'successToFailure', 'failureToSuccess']
[15:48:27] <lynxlynxlynx> gembot: notify list started
[15:48:27] <gembot> The following events are being notified: ['started', 'exception', 'successToFailure', 'failureToSuccess']
[15:48:30] <lynxlynxlynx> bah
[15:48:34] <lynxlynxlynx> gembot: notify off
[15:48:34] <gembot> The following events are being notified: []
[15:49:09] <lynxlynxlynx> anyway, if you want to hack it there, best to talk to tomprince, since he also works on that project
[16:00:03] <Avenger> yeah, i got the source, but it seems more difficult than i imagined :)
[16:01:54] <Avenger> gembot: What happen ?
[16:01:54] <gembot> Somebody set up us the bomb.
[16:02:02] <Avenger> ok... they got time for this...
[16:02:19] <wjp> I still think we should consider stealing scummvm's custom buildbot reporter, which is much nicer
[16:02:20] <lynxlynxlynx> i'm losing hair over the iwd2 inventory bugs
[16:02:46] <Avenger> well, i don't mind any solution that works :)
[16:03:07] <Avenger> just don't make a bot that triggers another bot :D
[16:03:17] <Avenger> and they start talking to each other
[16:03:46] <lynxlynxlynx> why does GemRB_GetSlotItem re-set the Slot for example? Without that, it would work ok in iwd2 too
[16:03:47] <rocket_hamster> gembot> Somebody set up us the bomb. :D
[16:03:57] <rocket_hamster> somebody set up us the bomb
[16:04:08] <Avenger> What you say !!
[16:04:15] <Avenger> gembot: What you say !!
[16:04:16] <gembot> You have no chance to survive make your time.
[16:04:18] <gembot> HA HA HA HA ....
[16:04:39] <rocket_hamster> Main screen turn on
[16:07:15] <Avenger> gembot: list builders
[16:07:15] <gembot> Configured builders: cmake clang++ cmake g++-4.2 cmake g++-4.4 cmake g++-4.5 cmake g++-4.6 mingw32 msvc++6 nmake-msvc++10 nmake-msvc++6 osx-cmake osx-xcode-binary
[16:07:23] <Avenger> watch cmake clang++
[16:07:27] <Avenger> gembot: watch cmake clang++
[16:07:27] <gembot> try 'watch <builder>'
[16:07:30] <Avenger> grr
[16:07:33] <tomprince> Quotes.
[16:07:35] <Avenger> ahh
[16:08:17] <Avenger> tom: it could somehow separate them in that list
[16:08:37] <Avenger> maybe using quotes itself
[16:21:04] <rocket_hamster> gembot: Somebody set up us the bomb.
[16:21:15] <rocket_hamster> hmm how does it work?
[16:21:27] <tomprince> It has a handful of canned responses.
[16:25:03] <rocket_hamster> gembot: What you say !!
[16:25:03] <gembot> You have no chance to survive make your time.
[16:25:06] <gembot> HA HA HA HA ....
[16:25:08] <rocket_hamster> yes
[16:25:21] <edheldil_> gembot: all your base are belong to us
[16:25:24] <traveler> gembot: watch "cmake clang++"
[16:25:25] <gembot> there are no builds currently running
[16:25:28] <edheldil_> hmm
[16:26:09] <rocket_hamster> how do you make it run in windows, if lets say user doesnt have python interpreter installed?
[16:26:41] <traveler> probably windows has some python with it
[16:26:46] <traveler> just a guess
[16:27:23] <rocket_hamster> oww im not sure
[16:28:14] <Avenger> you need to install python
[16:28:16] <tomprince> We bundle python.
[16:28:35] <tomprince> Or, do you mena buildbot.
[16:28:42] <tomprince> Then, yes, you need to install python.
[16:29:05] <rocket_hamster> tomprince: i was talking about gemrb in general, and i know what i did wrong
[16:29:07] <rocket_hamster> gembot: Somebody set up us the bomb.
[16:29:25] <rocket_hamster> hmm
[16:30:10] <Avenger> gembot: It's You !!
[16:30:11] <gembot> How are you gentlemen !!
[16:30:13] <gembot> All your base are belong to us.
[16:30:16] <gembot> You are on the way to destruction.
[16:30:21] <rocket_hamster> :D
[16:30:51] <rocket_hamster> i think i see the point now, you have to provide line that is directly before what Cats say
[16:30:52] <rocket_hamster> says
[16:34:33] <rocket_hamster> btw do you know you can watch star wars via telnet?
[16:34:35] <rocket_hamster> telnet towel.blinkenlights.nl
[16:34:43] <rocket_hamster> its random but fun :)
[16:37:57] <Avenger> almost 3d
[16:38:16] <rocket_hamster> put on the glasses :D
[16:56:49] --> brada has joined #gemrb
[17:16:43] <-- Den_ has left IRC (Ping timeout: 245 seconds)
[17:23:42] <-- fuzzie has left IRC (Ping timeout: 252 seconds)
[18:30:09] <lynxlynxlynx> cool, another patch
[18:30:32] <lynxlynxlynx> Avenger: here?
[18:30:36] <Avenger> yes
[18:31:11] <lynxlynxlynx> i can't wrap my head around this slottype issue in iwd2
[18:31:38] <lynxlynxlynx> i can see the weapon slots are in pairs and start at 43
[18:32:14] <lynxlynxlynx> then, when looking it up from the guiscript, we force it through slottype translation and it ends up as 35
[18:32:23] <lynxlynxlynx> and of course nothing is found
[18:32:28] <Avenger> hmm
[18:32:52] <lynxlynxlynx> is the table correct? iirc 35 was the weapon slot in bg2
[18:33:03] <lynxlynxlynx> and the ordering matters a lot
[18:33:16] <lynxlynxlynx> or is just the input data the problem
[18:33:32] <Avenger> i don't know
[18:33:43] <Avenger> 35 is indeed the bg2 weapon slot
[18:33:50] <Avenger> and 43 is the iwd2 weapon slot
[18:33:59] <Avenger> but the engine handles this
[18:34:07] <Avenger> it stores the first weapon slot in a variable
[18:34:19] <lynxlynxlynx> SLOT_MELEE i guess
[18:34:22] <Avenger> yes
[18:35:10] <lynxlynxlynx> GemRB_GetSlotItem does the translation before the lookup, so that bit doesn't help
[18:35:12] <Avenger> it shouldn't be able to end up as 35 if 35 is not hardcoded
[18:35:26] <lynxlynxlynx> Slot = core->QuerySlot(Slot);
[18:35:43] <Avenger> probably it loads a bg2 slottype.2da?
[18:35:50] <lynxlynxlynx> why is this needed at all? are there really two enumerations of inventory slots?
[18:35:57] <Avenger> yes
[18:36:00] <lynxlynxlynx> iwd2 has its own with 43 in it
[18:36:05] <Avenger> there is a silly permutation
[18:36:20] <Avenger> the creators of IE were on crack, for sure
[18:36:38] <Avenger> the order in the creature file and the slot ids are different
[18:36:53] <lynxlynxlynx> eeh
[18:36:55] <-- brada has left IRC (Quit: brada)
[18:37:14] <lynxlynxlynx> ok, it's a data issue in this case then, i'm passing a value from the core back directly
[18:42:08] <lynxlynxlynx> cool
[18:42:27] <lynxlynxlynx> there are plenty other nasties in the inventory though
[18:42:55] <lynxlynxlynx> now it looks like all weapon sets are equipped sometimes
[18:43:16] <lynxlynxlynx> getting the shield bonus while using a bow
[18:43:25] <lynxlynxlynx> or the arrow bonus while using melee
[18:43:40] <lynxlynxlynx> plenty of wierd stuff :s
[18:43:45] <lynxlynxlynx> but one at a time
[18:44:05] <Avenger> hehe
[18:44:20] <Avenger> iwd2 compared the shield item with shield itemtypes
[18:44:30] <Avenger> what happens in bg2?
[18:44:59] <Avenger> i think we determine shield items by checking if the shield could be equipped in the weapon slot, if it isn't. then it is a shield
[18:45:07] <Avenger> This should be working in iwd2 as well
[18:45:44] <lynxlynxlynx> i just added some extra checks
[18:46:05] <Avenger> i don't know why you had to :D
[18:46:16] <lynxlynxlynx> we were consider any usable weapon in the shield slot as a weapon, even shields (for dualwielding)
[18:46:24] <lynxlynxlynx> or bows
[18:47:18] <Avenger> shields for dual wielding?
[18:47:29] <lynxlynxlynx> the best was when i had this printed: Ranged - Shield of the urchin
[18:47:59] <Avenger> but this affected bg2 too?
[18:48:46] <Avenger> i don't like that getshieldpenalty thing :(
[18:50:14] <Avenger> this is the 'official' way, already used: int weapon = core->CanUseItemType( SLOT_WEAPON, itm );
[18:50:31] <lynxlynxlynx> but shields can be used in weapon slots
[18:50:36] <Avenger> no
[18:50:44] <Avenger> shields can be used only in shield slots
[18:50:51] <Avenger> weapons can be used in weapon slots
[18:50:55] <Avenger> and shield slots
[18:51:06] <lynxlynxlynx> hmm
[18:51:13] <lynxlynxlynx> we do have other itemtype problems
[18:51:29] <Avenger> itemtype 2da is not set up correctly?
[18:51:31] <lynxlynxlynx> loading an old save i have plenty of stuff equipped that we think shouldn't be possible
[18:51:54] <lynxlynxlynx> eg. cloak of mystra on a mage
[18:52:00] --> brada has joined #gemrb
[18:52:23] <Avenger> hehe
[18:52:42] <lynxlynxlynx> or some other items in usually nonproblematic slots
[18:52:49] <lynxlynxlynx> bracers, neckalces
[18:52:53] <lynxlynxlynx> weapons
[18:53:09] <lynxlynxlynx> i guess i should look at them one at a time
[18:53:19] <Avenger> well, maybe the usability flags are not set up correctly for iwd2
[18:53:32] <Avenger> iwd2 has them in a different order than other games
[18:53:44] <Avenger> first classes, then alignment
[18:53:45] <lynxlynxlynx> or mage not being able to use wands or scrolls, top that
[18:54:56] <lynxlynxlynx> if you have time, please look at it
[18:56:19] <Avenger> well i looked at itemuse.2da and it seems ok, i will debug this
[18:57:33] <lynxlynxlynx> let me know if you need a save
[18:58:50] <Avenger> huh
[18:59:21] <Avenger> gemrb just exits for me :D
[19:00:10] <Avenger> you merged iwd2/guiopt i guess
[19:01:08] <Avenger> i had lots of garbage in guiscripts
[19:01:27] <Avenger> damn, still quits
[19:02:08] <Avenger> complains about some table
[19:03:04] <Avenger> i just don't see the log
[19:06:55] --> Cuvieronius has joined #gemrb
[19:10:01] <Avenger> haha never compiled a file logger
[19:11:13] <Avenger> [ResourceManager/ERROR]: Couldn't find '*.2da' // [Actor/FATAL]: TOHIT table for BARBARIAN does not exist!
[19:11:37] <lynxlynxlynx> that's odd
[19:12:14] <lynxlynxlynx> do you have unharcdoded set up?
[19:12:27] <lynxlynxlynx> barbarians are the first row and have the same table as fighters - BAATFGT
[19:12:49] <lynxlynxlynx> i suggest you remove your override and just checkout the remainder
[19:14:42] <Avenger> ok that worked
[19:15:32] <Avenger> when loading with two handed stuff like bow equipped, the shield slot is invisible
[19:15:59] <Avenger> actually, this isn't just the shield slot and not just two handed stuff
[19:16:15] <Avenger> luckily a reequip fixes them
[19:16:39] <Avenger> could be wrong saved data
[19:17:36] <lynxlynxlynx> i have saves from the original
[19:18:03] <Avenger> ok, now i'm at a point i can check the unusability :)
[19:18:12] <lynxlynxlynx> :)
[19:18:30] <lynxlynxlynx> bbiaf
[19:20:23] <Avenger> there is a disabled print you can use
[19:20:28] <Avenger> print("failed usability: itemvalue %d, stat %d, stat value %d", itemvalue, itemuse[i].stat, stat);
[19:20:32] <Avenger> in actor.cpp
[19:23:40] <Avenger> generic mage has no kit value?
[19:26:44] <Avenger> ok, that was a failure, i tried with a sorcerer :D
[19:31:03] <lynxlynxlynx> heh
[19:34:33] <Avenger> heh, spwi110z is not listed as identify item in iwd2?
[19:34:34] <Avenger> grr
[19:39:11] <lynxlynxlynx> we have SPWI110 in the table
[19:39:42] <lynxlynxlynx> just not the item one
[19:40:19] <lynxlynxlynx> yeah, the default has a different scroll name
[19:41:43] <Avenger> i added itemspec.2da
[19:42:06] <Avenger> the problem with iwd2 is: we use classes.2da for kitlist
[19:42:19] <Avenger> and GetKitIndex uses a different column
[19:43:14] <lynxlynxlynx> well, a lot of things rely on tha layout by now
[19:43:34] <lynxlynxlynx> as far as kits go, we have that function doubled on the guiscript side
[19:43:35] <Avenger> iwd2's kitlist is crap
[19:43:41] <Avenger> so we need our classes
[19:44:16] <lynxlynxlynx> the table is loaded now, so the kit value is easy to get
[19:44:42] <Avenger> yes, just GetKitIndex needs to be fixed
[19:45:49] <lynxlynxlynx> at least they removed that barbarian hack :)
[19:47:21] <Avenger> there is even a todo: //TODO: make kitlist column 6 stored internally
[19:47:27] <Avenger> that would be useful for me :D
[19:47:50] <lynxlynxlynx> you probably wrote that
[19:48:50] <lynxlynxlynx> i see, we load it on demand everywhere
[20:00:21] <Avenger> there is a circular reference in combatinfo.h and scriptable/actor.h
[20:00:43] <Avenger> ahh you moved combatinfo
[20:01:35] <Avenger> ok, that solved it
[20:01:54] <lynxlynxlynx> brada: you'll need to readjust the xcode project again, i wasn't sure what hashes to use
[20:02:30] <brada> i saw. i dont really want to do it until we work out a method for fixing the cmake build on mac
[20:02:50] <brada> because im tired of that alert dialog poping up in the middle of watching a movie :/
[20:03:49] <lynxlynxlynx> you mean the project generation or something else?
[20:04:09] <Avenger> haha
[20:04:45] <Avenger> lynx i see you sneaked some vector in
[20:04:55] <lynxlynxlynx> i use kdevelop and cmake does have a generator for it; but i haven't had to change anything since first generated, since it supports cmake itself
[20:05:05] <lynxlynxlynx> hah
[20:05:16] <brada> lynx: i mean how cmake wont copy the guiscripts and co into the app bundle
[20:05:37] <lynxlynxlynx> ah ok
[20:05:51] <brada> i sent you a patch some time ago but there was something you didnt like about it
[20:06:01] <brada> ive since forgotten what
[20:06:14] <lynxlynxlynx> let's reevaluate it
[20:07:19] <brada> ok ill try to conjure it up later on
[20:07:49] <lynxlynxlynx> Avenger: btw, the GetCombatDetails refactoring isn't finished yet. I was trying to fill in everything we need for iwd2 first and then see what can be streamlined
[20:08:20] <lynxlynxlynx> pretty silly long argument list
[20:12:44] <brada> iirc you had a problem with me removing all the cmake.txt files for guiscripts etc and replaced them with a function that maybe didnt work with older cmake? <-- im more or less guessing by looking at the patch
[20:12:59] <brada> btw is there any way to apply part of a git stash?
[20:13:18] <brada> i guess i can just discard the chuks i dont want out of it and pop it like normal
[20:18:40] <lynxlynxlynx> not that i am aware off
[20:19:11] <Avenger> hmm iwd2 uses a very lame kit system, worse than bg2
[20:19:17] <Avenger> totally hardcoded
[20:20:00] <Avenger> they don't use the kit index hack bg2 has, so they have limited kits (the kit value is the same as the usability)
[20:20:15] <Avenger> this makes it simple, but unmoddable
[20:21:04] <Avenger> i could simply skip the kit -> usability translation for iwd2
[20:21:22] <Avenger> i think i will just do that for now
[20:22:16] <brada> ill have to resolve some conflicts and add the unhardcoded directory to this patch first
[20:37:03] <brada> I want to compile a list of interface changes that need to be made for me or whoever to write and sdl2/opengl video driver
[20:37:22] <brada> we need to do something about the sprite2d class
[20:37:29] <brada> specifically its pixels member
[20:37:31] <Avenger> ok lynx, kit usability for iwd2 is fixed
[20:38:42] <lynxlynxlynx> yay
[20:39:08] <lynxlynxlynx> i'll check if i can remove that shield check then
[20:39:43] <brada> i dont know if each video dirver should have its own subclass of sprite2d or what tho
[20:39:56] <Avenger> that shield check should be converted to the other shield check i told you
[20:40:07] <Avenger> that wouldn't be affected by this
[20:40:24] <brada> hmm fuzzie is not arround today i see
[20:41:21] <Avenger> int weapon = core->CanUseItemType( SLOT_WEAPON, itm ); <-- this doesn't check any kit stuff. Just checks if an item could be used in a slot
[20:44:15] <brada> would it a problem if i make an SDL2 branch on sourceforge so that everybody can push and pull to it without screwing up the current implementation?
[20:44:30] <brada> not that i have anything to commit right now
[20:44:56] <lynxlynxlynx> Avenger: that's already in place, but wasn't enough now
[20:45:14] <lynxlynxlynx> brada: of course not
[21:06:38] --> fizzle has joined #gemrb
[21:07:15] <brada> bah fooled by fizzle again!
[21:08:57] <fizzle> sry! need new glasses? :)
[21:10:27] <lynxlynxlynx> i think the sprite discussion cropped up before
[21:10:44] <lynxlynxlynx> try searching via our online logs
[21:11:10] <lynxlynxlynx> Avenger: fix confirmed
[21:11:45] * wjp looks up. Sprites?
[21:11:52] <brada> heh yeah sprites
[21:12:05] <Avenger> wjp you got a watch on 'sprite' ?
[21:12:10] <wjp> no :-)
[21:12:17] <brada> well you should!
[21:12:18] <edheldil_> pah, just finished pulling a fridge from a kitchen locker due to door not closing properly,half an hour fiddling with the screws and door still not closing, finding one hinge is broken, replacing it for one taken from other locker and putting the fridge back
[21:13:28] <lynxlynxlynx> and heh, iwd2 fists are considered 2h weapons
[21:13:40] <brada> about the sprites im just wanting input on dealing with sprite pixels since you cant have direct access to texture data
[21:13:43] <edheldil_> makes sense
[21:14:20] <brada> the complicated bit i guess is the bam sprites
[21:14:37] <brada> since something needs to hold them apparently
[21:14:38] <wjp> brada: the vptr field is for internal use by the video driver
[21:14:50] <brada> yeah i know
[21:14:54] <wjp> brada: we don't currently subclass sprite
[21:15:00] <wjp> so why ask? :-)
[21:15:48] <brada> would you say that using vptr for bam sprites to store a struct to track pixels in addition to what it has now?
[21:15:53] <brada> is a good idea?
[21:16:59] <brada> and that raises the question of how to access pixels when we need to
[21:17:09] <brada> do we pass the sprite to the video driver and let it do the work
[21:17:27] <brada> or have subclasses that know the internals?
[21:17:49] <wjp> pass to the video driver
[21:18:00] <brada> ok
[21:18:35] <brada> and you have no problem with me using vptr in that manner described above?
[21:19:48] <wjp> hm
[21:19:51] <tomprince> I'd say we should subclass sprite.
[21:20:01] <wjp> maybe we shouldn't be using vptr for the BAM internals
[21:20:25] <brada> tomprince: thats the course id normally take as well
[21:20:47] <brada> but im tired of doing work and posting patches only to hear that i should have done it a diffrent way :p
[21:21:00] <brada> so im asking for the design decisions up front this time!
[21:21:01] <tomprince> I wrote some code doing that, quite some time ago, but I don't know what happened to it. (It might be in my github fork)
[21:21:23] <tomprince> brada: Certainly asking up front is good.
[21:21:48] <tomprince> My input is that you should refactor Sprite2D to have subclasses.
[21:22:42] <brada> i think that is best as well
[21:23:11] <brada> im all ears to contrary arguments, however
[21:23:16] <tomprince> wjp: ?
[21:23:28] <wjp> no objections
[21:23:35] <wjp> I'm all for getting rid of the void*
[21:23:47] <brada> cool
[21:23:57] <brada> ill take that route then
[21:24:01] <wjp> but please don't introduce multiple inheritance to handle the bam/sdl20 cases
[21:24:19] <brada> with sdl2 i have disabled bam support
[21:24:23] <brada> for simplicity :p
[21:24:27] <wjp> uh
[21:24:35] <brada> what?
[21:24:54] <wjp> I guess you don't mean that literally :-)
[21:25:09] <brada> i mean SupportsBAMSprites returns false
[21:25:36] <brada> not as of now, but i mean with the work im doing now
[21:26:20] <brada> so the bam sprite class will be difined once at the SDLVideo level
[21:26:30] <wjp> ah, right, BAMImporter already has code to convert to bitmaps
[21:26:35] <brada> yup :p
[21:30:43] <brada> i also need a plan for VideoXXX::GetPalette
[21:31:23] <brada> i guess just generate one on the fly from the palette texture?
[21:33:53] <brada> I would think it might be better to have subclasses the same as sprites but it seems like palettes are a bit more ensconced in the core code
[21:35:23] <brada> we would have to move the palette construction to the video driver, but i only want to do that if its a Good Thing
[21:38:27] <wjp> how are you handling palettes when rendering?
[21:39:18] <brada> i assume you are asking about with SDL2?
[21:39:22] <wjp> yes
[21:39:30] <brada> http://www.opengl.org/wiki/Common_Mistakes#Paletted_textures
[21:39:35] <brada> pretty much that way
[21:46:37] <wjp> and how do you manage/cache/handle the required color table textures?
[21:47:25] <wjp> (and are there any performance issues with changing these regularly?)
[21:55:08] <brada> wjp: i havent actually done anything yet
[21:55:14] <brada> just trying to plan what i need to do
[21:55:30] <brada> maybe ill revisit this after doing some simple test code
[22:08:26] <brada> im reasonably confidant i will be able to do something useful with sdl2 :p
[22:09:15] <brada> tho we may have to render GameControl in software for a time until somebody with more opengl know-how has the time to write some shaders
[22:34:18] <lynxlynxlynx> Avenger: one other question, completely unrelated
[22:34:27] <Avenger> ?
[22:34:45] <lynxlynxlynx> shouldn't action ids mentioned in gemact.ids override the actions from the original?
[22:35:18] <Avenger> yes
[22:35:25] <Avenger> they should
[22:35:44] <Avenger> are there any collisions?
[22:35:45] <lynxlynxlynx> hmm, needs a deeper look then
[22:35:52] <lynxlynxlynx> i wanted to introduce one
[22:36:14] <Avenger> why
[22:36:18] <lynxlynxlynx> turns out bg1's ReallyForceSpell works for dead casters too - like ReallyForceSpellDead in bg2
[22:36:28] <Avenger> ah i see
[22:36:29] <lynxlynxlynx> traveler found this bug in totsc
[22:36:36] <Avenger> crappy engine changes
[22:36:42] <lynxlynxlynx> so we just tried an override, but it didn't help
[22:37:03] <Avenger> i think you could simply check for a GF
[22:37:07] <lynxlynxlynx> removing AF_ALIVE did, so that's clearly the right approach
[22:37:08] <Avenger> in reallyforcespell
[22:37:19] <lynxlynxlynx> that would be much uglier
[22:37:26] <Avenger> hmm, yes, af_alive too
[22:37:48] <lynxlynxlynx> AF_ALIVE is the only thing that is different between the two
[22:38:20] <Avenger> hmm
[22:38:22] <lynxlynxlynx> 181 ReallyForceSpellDead(O:Target,I:Spell*Spell) <-- this was the attempt, id of ReallyForceSpell, name of the other
[22:39:02] <Avenger> the problem is dialogs
[22:39:30] <lynxlynxlynx> this was in a script
[22:39:38] <Avenger> bg1 should still use "reallyforcespell" in dialog but it is our reallyforcespelldead
[22:40:00] <Avenger> well, not many dead dialogs anyway
[22:40:33] <lynxlynxlynx> yeah, this is from the phoenix guards, who either fireball and respawn or die
[22:40:42] <lynxlynxlynx> on death
[22:40:46] <Avenger> yes, in bgee we had to change their script
[22:40:50] <Avenger> to reallyforcespelldead
[22:41:21] <lynxlynxlynx> figures
[22:41:48] <lynxlynxlynx> aren't dialogs also compiled btw?
[22:42:01] <Avenger> the dialogs are compiled by the engine
[22:42:15] <Avenger> when it executes the actions
[22:42:22] <Avenger> quite silly :)
[22:42:58] <Avenger> i don't know why didn't they used the same format for dialog and script
[22:43:35] <Avenger> this delayed compiling just causes problems with ids files
[22:44:08] <tomprince> Only for modders ...
[22:44:23] <Avenger> *cough*
[22:45:08] <Avenger> only for people who build upon an existing dataset
[22:45:22] <Avenger> for example: dltc was unaffected :) but bgee is affected
[22:49:57] <lynxlynxlynx> wow, they continue to amaze me
[23:05:03] --> fuzzie has joined #gemrb
[23:05:49] <rocket_hamster> hey guys what do you think? https://github.com/miartad/gemrb-config
[23:06:27] <rocket_hamster> first draft
[23:06:33] <tomprince> python3 :(
[23:06:37] <rocket_hamster> problem?
[23:06:45] <rocket_hamster> i thought gemrb uses python3
[23:06:52] <lynxlynxlynx> nope
[23:06:55] <rocket_hamster> damn
[23:07:27] <rocket_hamster> now this is what i call epic fail
[23:07:28] <rocket_hamster> :D
[23:08:50] <tomprince> There is some support code for python3 that is probably broken.
[23:08:59] <tomprince> But python2 is the supported config.
[23:10:08] <brada> it would be better imo to make gemrb work without a cfg then to generate one
[23:10:22] <lynxlynxlynx> nice effort though, especially for someone who dislikes python
[23:10:27] <brada> indeed
[23:10:57] <lynxlynxlynx> you mean a in-core widgetry on startup, don't you?
[23:11:52] <brada> that would probably be the best aproach
[23:12:04] <lynxlynxlynx> my practical comment would be to read the config types from our man page and use the provided config sample directly, so there's one file less to maintain
[23:13:05] <lynxlynxlynx> we had a Config.cpp long ago, but it did practically zero, was unused and then removed
[23:13:55] <lynxlynxlynx> rocket_hamster: try running it with python2, there's likely less complications than you think
[23:14:32] <rocket_hamster> from what ive gathered in past few days porting to python should be just a matter of renaming some functions and enumerations
[23:14:36] <lynxlynxlynx> the readme should also mention you need pygtk
[23:15:01] <rocket_hamster> i thought python3-gobject is substitute for it
[23:15:06] <lynxlynxlynx> it's not about porting, there's even a script to do the chory part of it, it's about availability of python3
[23:15:22] <lynxlynxlynx> oh
[23:15:36] <lynxlynxlynx> ok
[23:15:45] <rocket_hamster> it was just moved to different namespace and renamed from gtk to Gtk
[23:15:53] <rocket_hamster> everything else is very much the same
[23:18:48] <rocket_hamster> python2 or python2.7?
[23:19:06] <brada> i think we are supposed to be compatible all the way back to 2.5
[23:20:05] <brada> but i dont know what platforms dont have 2.7 readily available
[23:20:08] <lynxlynxlynx> that can be changed though if really needed
[23:20:10] <lynxlynxlynx> yeah
[23:20:40] <brada> 2.7 ships with mac os since version 10.7 and the ios build uses 2.7 already
[23:20:44] <brada> dont know about android
[23:20:53] <brada> or various linux distros
[23:21:04] <brada> i imagine for linux 2.7 is the standard
[23:21:54] <fuzzie> i assume android doesn't ship it by default
[23:22:48] <fuzzie> i seem to have a 2.6 binary on this device (but no libs at all), presumably samsung use it for something
[23:23:29] <fuzzie> but i assume there is no gtk on any of these platforms?
[23:26:33] <rocket_hamster> androii highly doubt
[23:26:37] <rocket_hamster> i*
[23:27:50] <fuzzie> it wouldn't be worth trying to ship someyhing like qt anyway, so doesn't really matter i think
[23:28:07] <brada> true
[23:28:23] <brada> all the more reason to work on a gui config :p
[23:29:10] <brada> just make a CHU and whatever else
[23:29:27] <fuzzie> would be on python too :)
[23:29:30] <fuzzie> in.
[23:30:00] <lynxlynxlynx> that could get ugly
[23:31:11] <brada> we dont have a lot of options without adding more dependancies
[23:31:12] <lynxlynxlynx> oh, isn't tk bundled with python as tkinter
[23:31:30] <lynxlynxlynx> not pretty, but you have a gui
[23:32:12] <brada> we could make some kind of ConfigurationContext class that we could construct using various means depending on the platform
[23:32:26] <brada> since im pretty sure tk wont work on iOS
[23:32:36] <brada> i had to strip an awful lot out to get it to build
[23:32:59] <lynxlynxlynx> right, people have problems even with core modules
[23:33:17] <lynxlynxlynx> i guess eating our own dog food really is the best way forward then
[23:33:44] <brada> its definately the most dependable across all platforms
[23:33:55] <lynxlynxlynx> a default gametype of let's say config and special-casing in init to not bork out
[23:34:02] <brada> tho i would be more than willing to write mac/ios guis if i had soemthing to wire them to
[23:34:39] <lynxlynxlynx> it's less of a maintenance burden if we have it only in one copy
[23:35:19] <brada> I agree, but nobody seems like they want to make teh necessary CHUs et al
[23:35:46] <lynxlynxlynx> i can
[23:35:52] <brada> but will you? :)
[23:36:13] <lynxlynxlynx> when needed :)
[23:36:20] <brada> even with that i dont see a good way to marry it to core atm
[23:36:35] <lynxlynxlynx> i'm in an iwd2 killing streak now, so don't want too many distractions
[23:36:41] <brada> keep it up
[23:36:56] <lynxlynxlynx> no need to marry it, once the configuration is done, we would just quit
[23:36:59] <rocket_hamster> it could be done in html5?
[23:37:13] <lynxlynxlynx> next startup is with the config already and everything is back to normal
[23:37:29] <lynxlynxlynx> rocket_hamster: do you want to implement a html5 compliant browser in gemrb?
[23:37:38] <traveler> remind me why gui is so important?
[23:38:00] <lynxlynxlynx> because we have mostly human users
[23:38:07] <brada> because cfg is a PITA
[23:38:09] <traveler> *i'm not being overly sarcastic i hope
[23:38:11] <traveler> bugger
[23:38:26] <brada> ask any of our android users :p
[23:38:34] <rocket_hamster> lynxlynxlynx: dunno what you said but it sounds scary traveler: i noticed some people had trouble setting up config recently
[23:38:53] <lynxlynxlynx> all. the. time.
[23:39:10] <lynxlynxlynx> one path to set, yet ...
[23:39:35] <rocket_hamster> btw c++ has some standard gui library
[23:40:02] <brada> o_O
[23:40:06] <brada> define "standard"
[23:40:21] <rocket_hamster> well from my impression it is part of standard toolkit
[23:40:38] <brada> but is it available on all our targets?
[23:40:40] <traveler> i just hope gemrb will not end up pulling in qt6.66 and gtk5 on top of MONO and moonlight one day
[23:40:56] <brada> we dont want to add any additional dependancies for this
[23:41:09] <brada> so i wouldnt worry about that traveler
[23:42:17] <traveler> ncurses? some question based script?
[23:42:31] <brada> how will that help android?
[23:42:38] <brada> not very "gui" either
[23:43:33] <traveler> i think that gemrb is not very popular on android in the first place
[23:43:46] <rocket_hamster> traveler: you have to be able to click it or drag and drop it :)
[23:43:53] <traveler> contrary to the configure problems it generates, yes
[23:43:57] <brada> part of the reason for that would be difficulty in getting it running
[23:44:01] <brada> plus poor performance
[23:44:12] <traveler> or the old gemrb in market?
[23:44:16] <brada> that too
[23:44:24] <lynxlynxlynx> btw, it looks like spaces are too small now
[23:44:33] <traveler> and phones not being terrible handy to really playing
[23:44:39] <lynxlynxlynx> a px or maybe two
[23:44:48] <brada> the only realistic gui config options imo are to roll our own CHU as lynx wants to do and/or make a config class that can be bound to either CHU or platform gui controls
[23:45:15] <brada> traveler: you are forgetting the plethora of tablets
[23:45:26] <brada> which are quite nice for gemrb
[23:45:31] <lynxlynxlynx> or go watch the dingoo video again
[23:45:37] <brada> except no android build uses our touch input :/
[23:45:47] <lynxlynxlynx> completely shitty controls and 320x screen
[23:46:01] <rocket_hamster> what is CHU?
[23:46:04] <lynxlynxlynx> yet it was ported and people are happy
[23:46:21] <lynxlynxlynx> what the games use for defining gui layouts
[23:46:30] <brada> anyway a zoom feature a la bg:ee will be easily done once i get a working sdl2 driver
[23:47:02] <brada> ^ i already managed a hacked proof of concept here
[23:47:50] <brada> that would make targeting phones more realistic
[23:48:01] <brada> ie not a complete waste :p
[23:50:08] <traveler> to be honest, i have myself bg2demo on phone... but only to brag :P
[23:51:08] <traveler> iirc last time i checked more recent -git apk than that from the market had some problems with running
[23:51:20] <traveler> but not that i cared so much to investigate further
[23:53:11] <brada> it would help if we had an android build slave
[23:53:33] <lynxlynxlynx> did you write to alex?
[23:53:36] <brada> no
[23:53:45] <brada> i should
[23:54:12] <brada> i havent had time for much gemrb lately and what time i have had i spent on opengl/sdl research
[23:56:58] <traveler> i don't really know how 'market' works, but maybe it would be beneficial to announce that some android 'market maintainer' is needed who actually cares about android to support recent market package?
[23:58:04] <traveler> as in taking time to actually track gemrb releases
[23:58:13] <lynxlynxlynx> we're actively looking for new people, though not as much as we could
[23:58:26] <lynxlynxlynx> it's not just about rebuilding on later code though
[23:58:31] <lynxlynxlynx> beholder was here the other day
[23:59:18] <lynxlynxlynx> uhh, look at the time