#gemrb@irc.freenode.net logs for 29 Aug 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:24:15] <-- tombhad-AC has left IRC ("Verlassend")
[00:49:00] <-- raevol has left IRC ("Leaving.")
[01:08:37] <-- Edheldil has left IRC (Remote closed the connection)
[06:29:52] --> Gekz has joined #GemRB
[06:51:46] --> Edheldil has joined #gemrb
[06:51:46] --- ChanServ gives channel operator status to Edheldil
[07:25:13] --> lynxlynxlynx has joined #gemrb
[07:25:14] --- ChanServ gives channel operator status to lynxlynxlynx
[07:44:31] --> raevol has joined #GemRB
[07:52:20] <-- Gekz has left IRC (Remote closed the connection)
[07:58:53] <fuzzie> morning
[08:01:09] <lynxlynxlynx> oj
[08:13:46] <fuzzie> huh, didn't take long for gemrb 0.5.1 to get into gentoo
[08:15:10] <lynxlynxlynx> nick ;)
[08:15:23] <fuzzie> no, i mean, someone actually committed it :)
[08:15:25] <lynxlynxlynx> i didn't know it was in the official tree though
[08:15:42] <fuzzie> gemrb-0.5.1: 27 Aug 2009: initial commit - bug #158336 with contributions from Nick White
[08:16:31] <fuzzie> perhaps "roughly completable" were the magic words :)
[09:02:40] --> Gekz has joined #GemRB
[09:03:02] <fuzzie> one thing i have discovered is that it's pointless for me to run areas that i don't remember well
[09:03:15] <fuzzie> because i think everything's working fine, and then i play it in the original game and it's completely different. meh.
[09:14:32] <Edheldil> hi
[09:14:43] <fuzzie> hi, Edheldil
[09:17:50] <wjp> morning
[09:22:45] <Gekz> fuzzie: is it difficult to install Linux on a G3 system?
[09:22:53] <fuzzie> nope
[09:27:43] <Gekz> how so
[09:29:26] <lynxlynxlynx> it sports a common architecture?
[09:29:37] <fuzzie> well, you just put in a disc and install?
[09:31:13] <Gekz> I see
[09:31:34] <fuzzie> am i being stupid and missing something?
[09:31:52] <wjp> the majority of the common linux applications and libraries is pretty portable
[09:39:27] <fuzzie> apart from the lack of Flash, it's pretty nice :)
[09:40:40] <wjp> is gnash making any progress?
[09:47:38] <Gekz> swfdec is
[09:50:56] <Gekz> fuzzie: apparently swfdec is on ppc :)
[09:50:59] <Gekz> you can have flash
[09:51:08] <Gekz> install swfdec and swfdec-mozilla
[09:51:16] <lynxlynxlynx> king of the universe!
[09:51:25] <Gekz> who
[09:51:31] <lynxlynxlynx> flash
[09:51:35] <Gekz> lol
[09:55:54] <fuzzie> i tried both swfdec and gnash on x86 recently, nothing worked :/
[09:56:48] <lynxlynxlynx> just like flash
[10:12:13] --> Avenger has joined #gemrb
[10:12:18] --- ChanServ gives channel operator status to Avenger
[10:12:57] <Avenger> anyone awake with some medium level python knowledge?
[10:13:19] <wjp> hi Avenger
[10:13:30] <Avenger> hi, are you? :)
[10:13:33] <wjp> maybe :-)
[10:13:46] <Avenger> ok, i got a problem with lists
[10:13:55] <fuzzie> hi Avenger
[10:14:46] <Avenger> i got a list i want to initialise with tuple elements, i know how big it will be, and i don't need to change it again, so i allocate it this way: ContCond = [0] * CondCount
[10:14:59] <Avenger> is this fine so far?
[10:15:15] <Avenger> tuple = (Table.GetValue (i, 0),Table.GetValue (i, 1) )
[10:15:17] <Avenger> ContCond[i] = tuple
[10:15:27] <Avenger> this works
[10:15:41] <Avenger> printing ContCond lists a pretty list with tuples inside
[10:15:53] <Avenger> but, how do i iterate the elements O_O
[10:16:12] <fuzzie> does 'for i in ContCond:' not work?
[10:16:19] <Avenger> i tried this: for i in len (ContCond):
[10:16:21] <Avenger> print i
[10:16:33] <Avenger> hmm i don't need len?
[10:16:47] <wjp> if you want to loop over indices, it's 'for i in range(len(ContCond)):'
[10:16:53] <Avenger> ahh i see
[10:16:55] <wjp> in that case i will be 0, 1, 2, ...
[10:17:09] <fuzzie> but you can simply do 'for i in ContCond:' and then 'i' will be every tuple
[10:17:15] <wjp> if you do 'for i in ContCond:' then i will be (some_value_0, some_value_1), ...
[10:17:27] <Avenger> hmm, maybe i don't need the index, yeah
[10:17:36] <Avenger> i missed the range :)
[10:18:28] <Avenger> tuples are indexed the same way as arrays, right? like tuple[0] ?
[10:18:32] <wjp> yes
[10:18:34] <fuzzie> yes
[10:18:46] <fuzzie> but if you can get away without indexing, that is rather nicer :)
[10:19:14] <Avenger> yes fuzzie, i will try that first
[10:19:44] <fuzzie> now i wonder what you're doing, i was looking into converting dialogs of party members on expansion move with some similar code :)
[10:19:54] <Avenger> contingency screen
[10:20:00] <Avenger> spell sequencer, contingency
[10:20:01] <fuzzie> ah! neat
[10:20:36] <lynxlynxlynx> cool
[10:20:49] <Avenger> it is complicated, but in a good way
[10:20:57] <Avenger> i mean, i think i can cope with it :)
[10:21:27] <Avenger> did you try movetoexpansion?
[10:23:17] <lynxlynxlynx> you mean the wmap change? not yet
[10:24:19] <fuzzie> it turns out that we don't do ToB dialog files anyway
[10:24:25] <Avenger> yes i added it last night, it seems to be working
[10:35:31] <Avenger> heh, our guiscript docs are quite incorrect
[10:36:14] <Avenger> i tried to find how to use textarea.Append :)
[10:36:55] <Avenger> getting there :)
[10:38:23] <wjp> the docstrings don't really describe metaclass usage
[10:41:49] <Avenger> the textfile attempts it
[10:42:16] <Avenger> but that was the lesser evil, the bigger one was that i need to set -1 as row, if i want a new row
[10:42:25] <Avenger> the doc said the exact opposite
[10:42:49] <Avenger> it's my fault, i just forgot how to use it :)
[10:46:21] --> barra_library has joined #gemrb
[11:38:12] <lynxlynxlynx> yes, the wmap changing works fine
[12:03:07] --> barra_away has joined #gemrb
[12:19:10] <-- barra_library has left IRC (Connection timed out)
[12:24:30] <fuzzie> grr, stupid casting glow code
[12:24:44] <fuzzie> having it in GetEffectBlock is such a mess
[12:25:38] <fuzzie> Avenger: does it matter in which order the casting glow is applied? can i just add it afterwards?
[12:26:11] <Avenger> probably no
[12:26:47] <Avenger> i mean, you can add it afterwards, in the same gametime
[12:26:53] <fuzzie> ok :)
[12:40:05] --> tombhadAC has joined #gemrb
[12:53:18] <fuzzie> meh, the casting glow timing still isn't good, even fixed
[12:55:05] <Gekz> I am a bad person.
[12:55:07] <Gekz> How are you
[12:55:16] <fuzzie> you forgots the pythons, Gekz!
[12:55:22] <Gekz> I read
[12:55:30] <Gekz> IT WAS A PURPOSEFUL OMISSION
[12:55:32] <fuzzie> but it was quite amazing otherwise :)
[12:55:37] <Gekz> you dont deserve gemrb if you dont have Python installed
[12:55:39] <Gekz> xD
[12:56:19] <Gekz> fuzzie: I wonder if I could simply move the build env to linux
[12:56:23] <Gekz> and run it through wine
[12:56:39] <Gekz> sounds like a horrible hack
[12:56:43] <Gekz> but it might just work
[12:57:41] <Avenger> that would be insane to do
[12:58:00] <Gekz> insane hwo
[12:58:01] <Gekz> how*
[12:58:03] <Avenger> err, building is probably not insane
[12:58:11] <Avenger> i thought you want to play gemrb under wine
[12:58:21] <Gekz> neither of those things sound insane
[12:58:23] <Gekz> I'm going to try it now
[12:58:27] <Gekz> look what you made me do Avenger
[12:58:29] <Gekz> LOOK WHAT YOU DID
[12:58:30] <Gekz> lolo
[12:58:53] <Avenger> playing gemrb under wine? the bugs would just amplify each other
[12:59:00] <Gekz> haha
[12:59:03] <Gekz> it would make me laugh
[12:59:04] <Avenger> reaching critical mass
[12:59:06] <Avenger> boom
[12:59:09] <Gekz> it would esplode
[12:59:13] <Gekz> into many a core dump
[13:02:20] <fuzzie> ok, i propose http://fuzzie.org/nfs/gemrb/casting_time_fixes.txt if anyone feels like glancing through
[13:09:32] <Avenger> well test it a bit :) it is not like i could tell if it is wrong
[13:09:43] <Avenger> at a glance
[13:10:31] <fuzzie> well, it's a bit broken before, it's a bit broken after, difficult to tell :)
[13:10:45] <fuzzie> i think it's better, maybe i should see if lynx could try it
[13:10:55] <Avenger> you removed the need of supplying projectile in geteffectblock
[13:11:26] <Avenger> i guess now the cfb and the casting glow are applied separately
[13:11:56] <Avenger> i don't know why spellcast got a return value
[13:12:07] <fuzzie> because the action needs to know the duration of the cast
[13:12:50] <fuzzie> and, no, you still have to supply the projectile to GetEffectBlock, if you did before
[13:14:01] <Avenger> ahh the ext_index was removed
[13:14:14] <fuzzie> the ext_index isn't needed for AddCastingGlow either, i remove that too
[13:15:04] <fuzzie> but this is why i wondered if someone could look, because i'm not sure i changed it in a good way :)
[13:15:28] <Avenger> i don't see why it needs anymore
[13:15:40] <Avenger> casting glow doesn't need the ext_header if the duration is given
[13:15:50] <Avenger> it was only for the duration
[13:15:50] <fuzzie> yes, i removed that now, refresh the patch :)
[13:16:09] <fuzzie> i just forgot, i thought the compiler would warn me, but the assert stopped the warning :)
[13:16:42] <Avenger> i like that the casting glow is still in the same queue
[13:17:12] <Avenger> i wanted to avoid its separation, that's why it was hacked inside
[13:17:54] <Avenger> and what is this immediate flag? it would disable the casting glow?
[13:18:02] <Avenger> err instant flag
[13:18:05] <fuzzie> well, i'm not sure whether it does or not
[13:18:15] <fuzzie> it means "zero casting time"
[13:18:24] <Avenger> when is that used
[13:18:41] <fuzzie> in ReallyForceSpell
[13:18:50] <Avenger> maybe it has no casting glow
[13:19:06] <Avenger> not sure, though
[13:19:14] <Avenger> do we support the mentalspeed stat?
[13:19:27] <fuzzie> yes
[13:19:41] <Avenger> oh good, then only aura cleansing is needed
[13:19:51] <Avenger> aura cleansing: you cannot cast two spells in the same round
[13:19:54] <fuzzie> that would also be simple to add :)
[13:20:07] <fuzzie> but you can cast as many spells as you want in the same round at the moment
[13:20:23] <fuzzie> so someone's got to fix that first :)
[13:20:30] <Avenger> i mean, with aura cleansing it is how it would work :)
[13:20:48] <fuzzie> i thought IE_AURACLEANSING meant you *could* cast two spells in the same round
[13:20:55] <Avenger> yes, yes
[13:20:57] <Avenger> :)
[13:20:59] <fuzzie> ok, sorry! :)
[13:21:46] <fuzzie> i tried coding it, but i made the mistake of using the existing round variables
[13:22:19] <fuzzie> and it turns out they're handled a bit differently, so meh
[13:23:42] <Avenger> what is the syntax of a 'set' in Python?
[13:23:51] <Avenger> {} ?
[13:23:58] <CIA-22> gemrb: 03fuzzie * r7078 10/gemrb/trunk/gemrb/plugins/ (7 files in 2 dirs): spell casting time fixes
[13:24:19] <fuzzie> you want an actual set, or a dictionary?
[13:24:35] <Avenger> well, for exclusions a set is enough
[13:24:38] <fuzzie> i don't think there's any built-in syntax to declare sets
[13:25:03] <lynxlynxlynx> wow, we have aura dirtying now?
[13:25:05] <Avenger> so, i cannot initialise a list of sets?
[13:25:11] <fuzzie> you just pass a list to the Set constructor, Set(list)
[13:25:15] <fuzzie> like Set([1, 2, 3])
[13:25:17] <Avenger> no lynx: we have free aura cleansing
[13:25:19] <lynxlynxlynx> i can try the glow patch, but i'm not sure what to look for
[13:25:53] <Avenger> how do i make a list of 9 sets?
[13:25:56] <fuzzie> yes, me neither - it helps for the cases where casts were meant to be instant, and doesn't hurt the cutscenes i looked at, so i committed
[13:26:11] <Avenger> [ Set()] *9 ?
[13:26:54] <fuzzie> no
[13:27:01] <fuzzie> that way you end up with the same object, 9 times
[13:27:07] <lynxlynxlynx> the last commit added some waits
[13:27:35] <fuzzie> lynxlynxlynx: well, i removed SetWait from the spell code and moved it out into the action code, for now
[13:27:37] <lynxlynxlynx> list a = []; a.append(something)
[13:27:51] <lynxlynxlynx> oh
[13:27:54] <Avenger> lynx: i don't want 9 appends :)
[13:28:04] --> Enverex has joined #gemrb
[13:28:06] <lynxlynxlynx> loop?
[13:28:07] <Avenger> if i can avoid
[13:28:14] <Avenger> that's still 9 appends :P
[13:28:15] <fuzzie> in any case, sets are pure python i think, so we have to remember to bundle sets.py on Windows if we want to use them
[13:28:25] <Avenger> eeepp
[13:28:29] <Avenger> then no way
[13:28:39] <fuzzie> well, it's a few seconds to copy the py file over :)
[13:28:49] <Avenger> ok, what do you suggest, 9 have contingx.2da
[13:28:51] <fuzzie> i don't want to make your code harder
[13:28:54] <Avenger> i want to load its values
[13:29:06] <Avenger> 9 columns, variable number of resrefs in each
[13:29:32] <Avenger> so i need 9 sets which can tell me if a given resref is listed in it or no
[13:30:08] <fuzzie> you can simply do 'resref in list'
[13:30:41] <Avenger> you mean, i don't need to separate them into 9 separate lists?
[13:30:47] <Avenger> well, it would work
[13:30:58] <Avenger> not a big deal, it is still not too many
[13:31:05] <fuzzie> i mean, at leaat you don't need sets :)
[13:31:23] <Avenger> ok, i will use lists as sets :)
[13:31:40] <Avenger> so i can do [ []]*9 ?
[13:31:55] <Avenger> or that would still cause all 9 the same object
[13:32:07] <fuzzie> it still causes all 9 the same object
[13:32:12] <Avenger> python sucks :)
[13:32:21] <Avenger> how do i make them new objects
[13:32:23] <fuzzie> hehe, just use append :)
[13:32:34] <fuzzie> it is almost as fast, no big speed problem
[13:32:59] <Avenger> but there are some levels which got no spells in them
[13:33:33] <fuzzie> i mean, simply do 'for i in range(9): list.append([])'
[13:33:44] <Avenger> i see
[13:33:58] <Avenger> it sucks it cannot initialise compex types in a single shot
[13:35:02] <Gekz> haha
[13:35:04] <Gekz> cmake in wine
[13:35:08] <Gekz> is so incredibly slow xD
[13:41:36] <fuzzie> even with deepcopy, you can't do the [[]]*9 thing, because they really are identical
[13:42:03] <lynxlynxlynx> yeah, configure is slow too, usually you don't even notice it is compiling stuff for the tests
[13:44:18] <Avenger> ok, now i have another data structure, it is a list of spells, but each comes with a count, each spell is listed only once. I build this list from zero.
[13:44:40] <Avenger> i cannot use a mapping, because i want to keep the order
[13:44:54] <Avenger> so i cannot simply do a map[spell]++ like thing
[13:45:12] <Gekz> lol I am building in wine
[13:45:13] <Gekz> :o
[13:45:20] <fuzzie> well, you can keep a mapping and also an ordered list of the keys
[13:45:28] <Avenger> oh, hmm, ok
[13:45:40] <fuzzie> there is an ordered dictionary in python 3.1 but that is not so helpful
[13:45:49] <Avenger> how do i get an ordered list of the keys?
[13:46:28] <Gekz> Avenger: dict.keys().sort() ?
[13:46:31] <fuzzie> i mean, i thought you could keep a seperate list: every time you increase the count, do 'if not (spell in mylist): mylist.append(spell)' as well
[13:47:22] <Gekz> orderedlistofkeys = dict.keys().sort()
[13:47:39] <fuzzie> Gekz: but then that ends up in alphabetical order
[13:47:53] <fuzzie> while an ordered mapping would be in insertion order :)
[13:48:04] <Avenger> luckily alphabetical is fine
[13:48:09] <Gekz> oh
[13:48:12] <Gekz> you meant insertion order
[13:48:13] <Avenger> i think
[13:48:14] <fuzzie> oh, well, then you can just do what Gekz said :D
[13:48:16] <Gekz> I thought you meant proper order
[13:48:21] <Avenger> i don't know :)
[13:48:25] <Gekz> and what I Just said is redundant
[13:48:28] <Gekz> .keys() automatically sorts
[13:48:31] <Avenger> i don't know how the spells are listed in the contingency screen
[13:48:31] <Gekz> as I just found out
[13:48:35] <Gekz> in 2.6.2
[13:49:08] <Avenger> maybe they are not listed by resref...
[13:49:11] <Avenger> that would suck
[13:49:12] <Gekz> hey
[13:49:17] <Gekz> have any of you guys tested GemRB on ReactOS?
[13:49:24] <Avenger> no
[13:49:26] <fuzzie> no :)
[13:49:30] <Gekz> Gekz will.
[13:50:01] <Gekz> I'll even attempt to compile on it
[13:50:01] <Gekz> xD
[13:54:04] <Gekz> lol ReactOS installer is ugly
[13:59:07] <Avenger> ok, so names = SpellList.keys.sort() will work?
[13:59:26] <Gekz> keys() should sort it
[14:00:04] <fuzzie> keys() doesn't sort it, here
[14:00:18] <fuzzie> you have to do 'names = SpellList.keys()', 'names.sort()' or something
[14:00:26] <lynxlynxlynx> it would be a bit silly too
[14:00:27] <fuzzie> sort() is the worst python function ever
[14:01:41] <fuzzie> but i think there are just no promises about the order of keys()
[14:01:57] <Avenger> ok, spell is a string, spell[0] is a string too? or i need spell[0..0] ? or ?
[14:02:09] <fuzzie> yes, spell[0] is a single-char string
[14:02:35] <Avenger> bioware sucks, in contingx, they listed **** for empty resrefs, not the usual *
[14:03:00] <Gekz> well
[14:03:07] <Gekz> Avenger: names = SpellList.keys().sort() then
[14:03:13] <fuzzie> Gekz: sort() returns None
[14:03:38] <Avenger> yep, just names.sort()
[14:03:48] <Gekz> o.o
[14:03:51] <Gekz> stupid fucking
[14:03:52] <Gekz> -.-
[14:03:55] <fuzzie> like i said, worst python function ever
[14:03:58] <Gekz> yeah, I remember having to two-liner that
[14:04:10] <Gekz> it made me rage out the pants
[14:04:36] <lynxlynxlynx> i favour raging over import
[14:04:53] <Gekz> lolol
[14:04:59] <Gekz> ReactOS cant use vbox addons
[14:05:02] <Gekz> -.-
[14:05:03] <Gekz> fail
[14:05:11] <Gekz> why cant vbox emulate shared folders as usb devices
[14:05:12] <Gekz> come on
[14:06:13] <Avenger> cnt = len(names), right? ;)
[14:06:22] <fuzzie> let's hope so :)
[14:07:50] <Gekz> does len count 0 as the first?
[14:08:41] <lynxlynxlynx> sure
[14:09:37] <Avenger> len would be 0 for empty list :)
[14:10:01] <Avenger> cool, it is almost working
[14:10:13] <Avenger> i still wonder wtf find familiar does on the list
[14:10:49] <Avenger> the spell icons are not the right cycle, and the button font is not the nice small one, but it works
[14:14:41] <lynxlynxlynx> maybe the target option is the one restricting the spells
[14:15:09] <Avenger> the familiar is listed on the exclusions, it is some case problem
[14:15:11] <lynxlynxlynx> i don't know about the chain contingency and the best trigger, but the lower sequencer can only target the caster
[14:15:30] <lynxlynxlynx> ah, simpler :)
[14:18:30] <Avenger> hmm lowercase? s=lower(s) ? or s.lower() ?
[14:18:53] <Gekz> s.lower()
[14:19:01] <Gekz> s = s.lower()
[14:19:29] <Gekz> >>> "LOL".lower()
[14:19:30] <Gekz> lol
[14:29:16] <Gekz> its very difficult to get the files in reactos
[14:29:17] <Gekz> haha
[14:29:24] <Gekz> firefox keeps crashing
[14:29:33] <Gekz> and the only way to get stuff into it is through the internet
[14:29:33] <Gekz> -.-
[14:35:58] <Avenger> ok, another python question
[14:36:10] <Avenger> i have a list of things i don't want to put in a list
[14:36:14] <Avenger> so, i do this:
[14:36:17] <Avenger> if ms in Dummy:
[14:36:19] <Avenger> dummy[dummy.index(ms)]=None
[14:36:21] <Avenger> continue
[14:36:52] <Avenger> is this fine? or is there an easier solution
[14:37:06] <Avenger> i find it sad i need to do an 'in' and then an index
[14:37:40] <Avenger> oops and ignore the case on Dummy
[14:37:45] <Gekz> I dare say Python was an ill-guided choice xD
[14:37:56] <fuzzie> i think everything else would be more horrifying :)
[14:38:05] <fuzzie> i thought gemrb was crazy to use python for the gui when i first saw it
[14:38:06] <Gekz> Lua?
[14:38:21] <fuzzie> but then i tried Lua in another engine and, well, it's not so nice :/
[14:38:46] <fuzzie> Avenger: i don't know what you try to do there :(
[14:38:56] <Avenger> i like python, but the stuff i code now is something i would normally do in C
[14:39:01] <Avenger> hehe, you don't?
[14:39:08] <Avenger> ok lemme explain
[14:39:11] <fuzzie> you just try and remove it from the list? why not 'del'?
[14:39:17] <Avenger> dummy = [Spell1,Spell2,Spell3]
[14:39:22] <Avenger> ahh yes
[14:39:55] <fuzzie> but in any case, i think 'dummy.remove(ms)' is what you want
[14:39:56] <Avenger> that's not much better, but i could go with del
[14:39:58] <lynxlynxlynx> or "as"
[14:40:04] <Avenger> oh hmm
[14:40:06] <Avenger> ok
[14:40:40] <Avenger> you know, i prefer overwriting an array element than moving the whole crap around :) but these are just lists, right?
[14:41:19] <Avenger> and with 3 elements i could even make a phone call
[14:41:26] <Avenger> so it isn't a big deal
[14:41:27] <fuzzie> i think you worry a little too much about performance :)
[14:41:45] <Avenger> well i'm a c programmer
[14:41:48] <fuzzie> truly, the rendering stuff in the core makes python completely irrelevant
[14:41:56] <fuzzie> and makes all the rest of our inefficiencies mostly irrelevant, really
[14:43:04] <fuzzie> so someone coding an opengl backend would outweigh any python optimisations by, maybe a few tens of thousand of times :)
[14:43:12] <Avenger> i optimise subconsciously
[14:44:01] <Avenger> you say it would be possible to speed up the sdldriver?
[14:44:17] <fuzzie> sure, we just re-render everything right now, i think?
[14:44:31] <Avenger> not sure we rerender all
[14:44:34] <fuzzie> you could do all kinds of optimisations like dirty-rectangling, even if you didn't use opengl
[14:44:41] <fuzzie> but it's a lot of annoying work to do
[14:45:28] <fuzzie> might be interesting to do more profiling
[14:45:52] <Avenger> damn familiar is still on the list :(
[14:47:00] <Avenger> ahh hehe, i found why
[15:03:09] --> xrogaan has joined #gemrb
[15:03:47] <Gekz> ok
[15:03:47] <Gekz> guys
[15:03:54] <Gekz> you cant compile GemRB in ReactOS
[15:03:56] <Gekz> cmake dies
[15:04:03] <Gekz> it's not all bad
[15:04:09] <Gekz> because the one I compiled last night ran
[15:04:17] <Gekz> so I'm installing IWD now to see if it works with actual games
[15:04:22] <Gekz> if so, this is rather epic win
[15:04:35] <Enverex> Can't you just compile it for Windows and run the Windows version in ReactOS?
[15:04:43] <Gekz> Enverex: what did I just say
[15:04:53] <fuzzie> Enverex: that's no fun, is i think the point here :)
[15:05:22] <Enverex> Gekz, I mean without having to compile it inside ReactOS itself, just use an existing binary (who seriously compiles programs when running Windows?)
[15:05:32] <Gekz> Me!
[15:06:12] <fuzzie> someone's got to do the compiling sometime :(
[15:07:35] <Gekz> this time it was me
[15:07:37] <Gekz> with a sad face
[15:07:39] <Gekz> and a top hat
[15:07:59] <fuzzie> and a copy of Windows 7, which is cheating
[15:08:10] <fuzzie> you are banished from the island
[15:08:19] <Gekz> how is it cheating
[15:09:20] <Gekz> haha I dont think the installer can finish
[15:09:20] <Gekz> xD
[15:09:32] <Avenger> oh my, sorcerors are difficult
[15:09:43] <Avenger> sorcerors and their sequencers
[15:10:14] <fuzzie> gah, why do people keep on pointing CachePath at /var/cache/gemrb?!
[15:10:15] <Avenger> if you put one of the spells in a sequencer, all the other spells on that level need to be decreased :(
[15:11:20] <Avenger> i don't think they do it on their own
[15:11:20] <Gekz> haha what the hell
[15:11:24] <lynxlynxlynx> fuzzie: old config; no matter what we or they change in the default config, the user already has his own
[15:11:29] <Gekz> This application has failed to start because
[15:11:31] <Gekz> WAIT FOR IT
[15:11:36] <Gekz> MSVCR90.DLL was not found.
[15:11:48] <fuzzie> lynxlynxlynx: this is from the gentoo package
[15:11:54] <lynxlynxlynx> hehe
[15:12:03] <fuzzie> i only just got it removed from the ubuntu one :(
[15:12:30] <lynxlynxlynx> i don't have an account on their hive, njw isn't on freenode either now
[15:12:31] <fuzzie> i guess it is nick white's config
[15:12:37] <lynxlynxlynx> wjp: do you have one?
[15:12:48] <lynxlynxlynx> hive == bugzilla
[15:22:15] --> pupnik has joined #gemrb
[15:36:32] <Avenger> now i don't get those gui breakdowns
[15:36:34] <Avenger> odd
[15:36:34] <Gekz> right
[15:36:38] <Gekz> how can I _force_ null sound
[15:36:50] <Avenger> delayplugin for openal
[15:37:35] <fuzzie> Avenger: you manually saved a game first?
[15:37:55] <Avenger> hmm didn't save now
[15:37:59] <Gekz> Avenger: elaborate
[15:38:22] <Avenger> gekz: there is a delayplugin option, you should see an example in the cfg
[15:38:23] <-- pupnik_ has left IRC (Connection timed out)
[15:38:29] <Gekz> so
[15:38:40] <Gekz> libOpenALAudio
[15:38:46] <Avenger> so delay the openal plugin, that will let the nullsound load first
[15:39:28] <Gekz> OH MY GOD
[15:39:29] <Gekz> I GOT IT WORKING
[15:39:41] <Gekz> IWD in ReactOS
[15:39:47] <Gekz> IWD cant run natively in ReactOS
[15:39:50] <Gekz> This is epic win1
[15:40:44] <fuzzie> heh
[15:41:48] <Gekz> by god does it run slow
[15:41:54] <Gekz> but it runs nonetheless
[15:42:47] <Avenger> heh, my sorcerer cannot die
[15:42:55] <Avenger> any wound it gets, it heals back
[15:43:13] <fuzzie> a ctrl-y or two might fix that
[15:43:21] <Avenger> minsc doesn't fight in berserker mode
[15:45:43] <Gekz> http://superiorchicken.info/reactos-iwd.png
[15:45:45] <Gekz> http://superiorchicken.info/reactos-iwd-2.png
[15:45:46] <Avenger> good for playthroughs, i just drop a few skull traps
[15:46:46] <Gekz> http://superiorchicken.info/reactos-iwd-1.png
[15:46:48] <Avenger> the installer works?
[15:47:18] <Gekz> yep
[15:47:19] <Gekz> :)
[15:47:22] <Gekz> the game does not
[15:47:27] <Gekz> unless in GemRB
[15:48:13] <Avenger> without sound
[15:48:25] <Gekz> without sound, but that's virtualboxs issue
[15:49:57] <Gekz> hehe I'm taunting the #ReactOS guys with it
[15:49:57] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
[15:54:23] <Gekz> fuzzie: lol
[15:54:29] <Gekz> the ReactOS guys all shat bricks
[15:55:21] <lynxlynxlynx> fireball, vampyre, i can see why :)
[15:55:23] <Gekz> well I must sleep now :D
[15:55:25] <Gekz> bais
[15:55:31] <lynxlynxlynx> bb
[15:57:04] <-- Gekz has left IRC (Read error: 54 (Connection reset by peer))
[16:42:54] <Edheldil> lo!
[16:43:35] <fuzzie> hey
[16:50:14] <CIA-22> gemrb: 03avenger_teambg * r7079 10/gemrb/trunk/gemrb/includes/strrefs.h: new strrefs
[16:51:48] <CIA-22> gemrb: 03avenger_teambg * r7080 10/gemrb/trunk/gemrb/override/ (6 files in 6 dirs): new strrefs
[17:19:44] <Enverex> Text is still green I see :P
[17:28:17] <fuzzie> heh, we're a little distracted :)
[18:24:22] <-- Enverex has left IRC (Remote closed the connection)
[18:36:17] <CIA-22> gemrb: 03avenger_teambg * r7081 10/gemrb/trunk/gemrb/ (17 files in 11 dirs):
[18:36:17] <CIA-22> gemrb: implemented contingency and sequencer spells
[18:36:17] <CIA-22> gemrb: implemented LastSpell (SpellCast triggers could be done now)
[18:50:27] <fuzzie> i think perhaps LastSpell was not the difficult part of the triggers :-)
[18:52:27] <lynxlynxlynx> woah
[18:56:21] <Edheldil> what is contingency?
[18:58:55] <lynxlynxlynx> when the shit hits the fan, contingency is what saves you
[19:03:49] <wjp> it lets you automatically cast some spells when a certain condition (like having few hitpoints left) is met
[19:12:18] <fuzzie> one of the last missing gui bits, i think.
[19:22:48] --> edheldil_ has joined #GemRB
[19:24:58] <lynxlynxlynx> half of its gui is still missing though
[19:25:32] <edheldil_> iwd2's?
[19:25:43] <lynxlynxlynx> no, this contingency thing
[19:27:28] <lynxlynxlynx> and it always seems to target the caster :/
[19:29:02] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[19:33:50] --> Avenger has joined #gemrb
[19:33:57] --- ChanServ gives channel operator status to Avenger
[19:34:00] <Avenger> hello
[19:34:59] <Avenger> i still don't know what causes the gui breakdown, today i fought a lot, and there was not a single case
[19:48:15] <edheldil_> Avenger: [ [] for i in range (9) ]
[19:48:55] <edheldil_> and for index, item in enumerate (somelist): ...
[19:49:29] <edheldil_> and there's now also sorted(), which does not sort in place
[19:51:03] <fuzzie> we support 2.3+ or something, i think
[19:51:41] <fuzzie> [ [] for i in range (9) ] is neat, though!
[19:51:50] <edheldil_> do we really? Has anyone checked recently? :) (I did not)
[19:52:02] <fuzzie> i think i run it on 2.4
[19:52:11] --> pupnik_ has joined #gemrb
[19:52:21] <fuzzie> hm, README says 2.2+ :)
[19:52:21] <edheldil_> I like enumerate, it's such a nice idea
[19:53:12] <edheldil_> let's make it 2.5+ ;-)
[19:53:12] <fuzzie> for 2.2 you don't even have enumerate, heh
[19:53:21] <fuzzie> well, i would quite like you to not break my 2.4 :)
[19:54:13] <wjp> enumerate seems to work in 2.4.4 for me
[19:54:23] <fuzzie> yes, enumerate is 2.3
[19:54:59] <fuzzie> sorted() is not in 2.5, and i don't see it in 2.6 docs?
[19:55:31] <Avenger> its fine to skip 2.2, it was old when we started gemrb :)
[19:55:46] <Avenger> but lets not jump too ahead
[19:57:17] <Avenger> so, what are you people doing?
[19:57:48] <fuzzie> well, i am testing some more action bits, and then i want to remove WaitCounter and fix more action framework stuff
[19:58:33] <Avenger> i wonder who broke the stance/animations
[19:58:53] <fuzzie> well, the combat stance code is completely crazy right now, in any case
[19:59:10] <fuzzie> i don't see how it ever worked
[19:59:17] <fuzzie> if it's not combat stances i don't know :)
[20:00:35] <fuzzie> it looks like you could require python 2.5 for gemrb without many compatibility problems
[20:01:52] <fuzzie> Avenger: do you want anything particular looked into? and i fear the windowing code, so not that :)
[20:02:14] <Avenger> well, what about curved path magic missile ?
[20:02:25] <Avenger> that's not scripting at least
[20:02:45] <-- pupnik has left IRC (Read error: 110 (Connection timed out))
[20:02:59] <Avenger> i'm testing HoW now, i want to reach the avalanche at least
[20:03:04] <Avenger> i did it earlier
[20:03:47] <Avenger> the bugs quest and the orc cave was working, i hope they remained working
[20:06:31] <Avenger> hmm, i think red feet circle should never be covered out, right?
[20:14:31] <Avenger> fighting animations suck :)
[20:14:46] <Avenger> the constant hitting is gone, but it was like that in the original
[20:15:21] <Avenger> if you fight the animation loops on hitting, even without real attack rolls
[20:31:27] <fuzzie> covered out?
[20:31:46] <fuzzie> and, yes, the combat animations should fill in the gaps
[20:32:03] <fuzzie> you can't just loop the animation though, then you don't get the actual attacks right
[20:34:45] <fuzzie> but if you get a savegame before the avalanche, i would like it :)
[20:35:27] <Avenger> you NEVER get the actual attacks right :)
[20:35:34] <Avenger> not even in Dragon Age
[20:35:47] <fuzzie> oh? it seemed pretty correct, at least in bg1
[20:35:54] <Avenger> no, never
[20:36:09] <Avenger> you can read about it on forums :)
[20:36:29] <fuzzie> heh, the forums are full of lies :(
[20:37:15] <Avenger> well, i read about dragon age, only. But if they couldn't do this in 2009, with 3d animation technology, then what do we want :>
[20:37:21] <fuzzie> well
[20:37:47] <fuzzie> whenever they talk about Dragon Age they seem to go "well, we hope to get somewhere near Baldur's Gate 2 again", so .. :P
[20:38:08] <Avenger> hehe
[20:38:18] <fuzzie> but i'll take a look
[20:38:20] <Avenger> btw, lynx tangled these scripts quite much
[20:38:43] <Avenger> it would need someone with a more lawful alignment :)
[20:39:08] <Avenger> GetActorClassTitle in particular
[20:39:27] <Avenger> it started with returning a numeric value, an strref
[20:39:37] <Avenger> and some parts of it still think they will want to return numeric
[20:40:04] <fuzzie> doesn't it work fine with either a strref or a string?
[20:40:19] <Avenger> what's worse, other code thinks too
[20:40:35] <Avenger> File "./GUIScripts/iwd/GUIREC.py", line 242, in UpdateRecordsWindow
[20:40:37] <Avenger> ClassTitle = GemRB.GetString (GetActorClassTitle (pc) )
[20:40:39] <Avenger> AttributeError: GetString(strref[,flags]) => string
[20:40:40] <Avenger> this is HoW
[20:40:43] <fuzzie> ah
[20:40:45] <Avenger> record window is broken right now :)
[20:40:53] <fuzzie> well, i guess that needs fixing
[20:41:01] <Avenger> i do a little synchronising
[20:41:11] <fuzzie> well, be careful when synchronising :)
[20:41:12] <Avenger> meh, i always do this
[20:41:33] <fuzzie> we try to slowly refactor the code to be shared, rather than just copying it to every game
[20:51:47] <Avenger> ok i kicked it to stand up and walk, but it is still a bit ill
[20:51:55] <CIA-22> gemrb: 03avenger_teambg * r7082 10/gemrb/trunk/gemrb/ (6 files in 4 dirs): fixed IWD records window
[20:52:11] <Avenger> yes i saw that
[20:52:42] <Avenger> i just patch the old code to stay compatible with the common changes
[20:53:00] <Avenger> it is not a problem if we introduce kits to iwd or bg1 :)
[20:53:11] <Avenger> they actually had those mage schools
[20:53:38] <Avenger> the kits in bg2 are just hacked on it
[21:26:26] --> dawid has joined #GemRb
[21:30:56] <-- dawid has left #GemRb ("Leaving.")
[21:35:55] <-- |Cable| has left IRC (Remote closed the connection)
[21:37:19] --> |Cable| has joined #gemrb
[22:04:09] <fuzzie> Avenger: what do you think of just assigning everything unique global ids, and then having a seperate function to search everything (or optionally, only a certain type of scriptable) which is only used when necessary?
[22:04:45] <fuzzie> it seems the simplest approach, and i see very few places where we'd actually need to search everything
[22:05:00] <Avenger> i said yesterday, i would prefer some order with the global ids, like the high word would be the type
[22:05:22] <fuzzie> well, an index turns out to not work very well
[22:05:23] <Avenger> i don't mind if you assign some random value to the low word of non-actors, but i think index is good enough
[22:05:37] <fuzzie> because then i have to keep track of area somehow
[22:05:49] <fuzzie> but the high word being the type is easy
[22:06:20] <Avenger> but how do you think about the id, a running integer? or what?
[22:06:36] <fuzzie> well, i just thought i'd do them like the current actor ids, we simply add one every time we need a new id
[22:07:01] <Avenger> the actor ids are area specific, i think
[22:07:08] <fuzzie> the globalID is global
[22:07:28] <fuzzie> it really just adds 1 to a global variable every time it needs one, and uses that
[22:08:15] <fuzzie> i did notice i had to add a function to Game to actually obtain an actor from that ID
[22:08:48] <Avenger> well, i wish i know how this is done in the original
[22:09:13] <fuzzie> well, it's certainly not area-specific in the original
[22:10:18] <fuzzie> and from what everyone said about the code, a single global ID is enough to identify any object?
[22:11:20] <fuzzie> but i don't think it matters how we save them internally, i thought the original engine doesn't care?
[22:11:26] <Avenger> yes, it is a single dword on any scriptable objects 0x30 offset
[22:11:32] <fuzzie> i just want rid of the pointers, anyway
[22:12:24] <Avenger> i stil don't know how this number is chosen
[22:12:42] <fuzzie> the numbers aren't just consecutive?
[22:12:56] <Avenger> ...
[22:13:06] <fuzzie> well, i don't see any of the structs :)
[22:13:14] <fuzzie> so i don't know what the examples look like
[22:13:31] <Avenger> you can see actor global ids in saved game areas
[22:13:45] <Avenger> and that's why i'm a bit helpless
[22:13:57] <Avenger> iirc, they are two identical words in saved games
[22:14:37] <fuzzie> iesdp claims they're 'global' and 'local' values
[22:14:55] <Avenger> yes, that's why i made ours similar
[22:15:00] <fuzzie> but if they're two words in the save, and a dword in the struct, that seems stranger
[22:15:08] <Avenger> but for the actors i saw in saved games, they are equal
[22:15:22] <fuzzie> and my first thought would be that the save code is simply writing junk
[22:15:33] <Avenger> that is a possibility
[22:15:53] <Avenger> it is completely ignored on load, ithink
[22:15:59] <fuzzie> but presumably it is possible to break in a living-only action when it runs and peek at 0x30?
[22:16:18] <Avenger> yes it is very possible
[22:16:44] <Avenger> never looked at it , though
[22:17:06] <Avenger> but i will
[22:17:08] <Avenger> :)
[22:17:15] <Avenger> ok, i have to sleep now
[22:17:18] <Avenger> see you tomorrow
[22:17:25] <fuzzie> ninight!
[22:17:26] <-- Avenger has left IRC ("bye!")
[23:04:48] <-- barra_away has left IRC ("Verlassend")
[23:35:44] <-- xrogaan has left IRC ("Why ?")
[23:46:22] --> Gekz has joined #GemRB