#gemrb@irc.freenode.net logs for 27 Feb 2015 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[01:04:12] --> Eli2_ has joined #gemrb
[01:06:26] <-- Eli2 has left IRC (Ping timeout: 265 seconds)
[01:08:13] <-- fireflame has left IRC (Ping timeout: 246 seconds)
[02:20:11] --> turtleman_ has joined #gemrb
[02:53:56] <-- turtleman_ has left IRC (Ping timeout: 256 seconds)
[06:47:24] <-- Lightkey has left IRC (Ping timeout: 256 seconds)
[07:00:04] --> Lightkey has joined #gemrb
[11:40:05] <-- edheldil has left IRC (Remote host closed the connection)
[11:41:07] --> edheldil has joined #gemrb
[11:41:07] --- ChanServ gives channel operator status to edheldil
[13:35:46] <-- Pepelka has left IRC (Remote host closed the connection)
[13:43:48] --> turtleman_ has joined #gemrb
[14:06:47] <-- Drakkar has left IRC (Ping timeout: 246 seconds)
[14:09:57] --> Drakkar has joined #gemrb
[14:51:59] <-- turtleman_ has left IRC (Ping timeout: 250 seconds)
[16:15:47] --> brada has joined #gemrb
[16:17:13] --> Seniorita has joined #gemrb
[16:21:07] <brada> think im finally going to get around to TA editing capabilities this weekend
[16:22:19] --> turtleman_ has joined #gemrb
[16:25:14] <lynxlynxlynx> cool
[16:25:27] <lynxlynxlynx> don't forget about the iwd2 crashing though
[16:25:57] <brada> of course :)
[16:26:13] <lynxlynxlynx> i should add it to the test suite
[16:26:49] <brada> always helpful
[17:42:01] <-- turtleman_ has left IRC (Ping timeout: 250 seconds)
[18:00:42] --> kpedersen has joined #gemrb
[18:03:37] <-- kpederse1 has left IRC (Ping timeout: 245 seconds)
[18:23:26] --> Crevs_Daak has joined #gemrb
[18:30:44] --> turtleman_ has joined #gemrb
[18:40:29] --> kingron has joined #gemrb
[18:58:06] <-- turtleman_ has left IRC (Ping timeout: 252 seconds)
[19:31:57] <-- brada has left IRC (Quit: brada)
[19:33:04] --> brada has joined #gemrb
[19:43:49] <wjp> the Linux versions of BG/BG2/IWD EE on GOG seem to be working nicely
[20:04:21] --> kpederse1 has joined #gemrb
[20:07:02] <-- kpedersen has left IRC (Ping timeout: 252 seconds)
[20:15:27] <-- kingron has left IRC (Remote host closed the connection)
[20:34:02] <-- Crevs_Daak has left IRC (Ping timeout: 245 seconds)
[20:39:23] --> Crevs_Daak has joined #gemrb
[21:37:27] <lynxlynxlynx> ohai
[21:37:46] <lynxlynxlynx> didn't notice you came in before Crevs_Daak
[21:42:17] <Crevs_Daak> hey there
[21:42:20] <Crevs_Daak> no worries
[21:44:57] <Crevs_Daak> do you think it's safe to use s/^.*NumInParty(6)/NumInPartyGT(5)/g ??
[21:45:35] <lynxlynxlynx> hmm
[21:45:41] <Crevs_Daak> because it'll break checks to see if the party's full
[21:46:08] <lynxlynxlynx> for that purposes the chosen number would work fine
[21:46:17] <lynxlynxlynx> no need to change the trigger type
[21:46:36] <Crevs_Daak> OK
[21:46:39] <lynxlynxlynx> but if a script checks for different sizes, we'll have to cheat
[21:46:58] <lynxlynxlynx> a proper solution would be to hand change them, but that's not something you can automate
[21:47:34] <Crevs_Daak> also it will make the game easier for any spawns checking for real party size instead of L/GT
[21:47:56] <lynxlynxlynx> uff, i just realised i forgot something - only gemrb supports up to player10, we'll need to modify weidu for non-dialog scripts to be compilable
[21:48:31] <lynxlynxlynx> summons don't count for party
[21:48:31] <Crevs_Daak> I believe WeiDU takes the stuff from the .ids files
[21:48:53] <lynxlynxlynx> yes, overrides will do
[21:49:56] <lynxlynxlynx> my current script relies on a first semi manual pass, iow not scalable at all
[21:50:15] <lynxlynxlynx> so we just have to use much smarter regexes
[21:51:03] <Crevs_Daak> ok
[21:51:48] <Crevs_Daak> should we put the regexp on a perl script using a shitload of if-statements?
[21:53:05] <lynxlynxlynx> i'm not proficient in perl, but the language doesn't matter much
[21:53:36] <Crevs_Daak> ok
[21:53:36] <lynxlynxlynx> just some escaping needs to be done to convert, as sed uses non-perl regex by default
[21:54:41] <lynxlynxlynx> but yeah, it's the actual model that will be most of the work
[21:55:24] <lynxlynxlynx> i have never used lookahead and its reverse, except manually in sed
[21:55:31] <lynxlynxlynx> nice opportunity to learn that
[21:55:49] <Crevs_Daak> yup
[21:56:16] <Crevs_Daak> a sed script re-making the regexps would be easier than doing it manually
[21:56:29] <Crevs_Daak> (everything's easier that doing that manually)
[21:56:34] <lynxlynxlynx> the whole point is to automate it
[21:56:42] <lynxlynxlynx> now it's a bunch sed scripts
[21:56:59] <lynxlynxlynx> awk only for deduping and plenty of diff parsing
[21:57:05] <lynxlynxlynx> lame, but kinda worked
[21:57:11] <Crevs_Daak> i can make the weidu code in some minutes
[21:57:18] <Crevs_Daak> brb
[21:57:27] --- Crevs_Daak is now known as Crevs_away_5_min
[22:01:42] <lynxlynxlynx> lynxlynx.info/ie/how.baf.tar.bz2 <-- here's a dump of the iwd:how script files, if you want a test set
[22:01:58] <lynxlynxlynx> i thinking i damaged by bg ones while playing around and need to regenerate them
[22:05:41] --- Crevs_away_5_min is now known as CrevsDaak
[22:05:55] <CrevsDaak> today I extracted all og my modded BGT's scripts
[22:06:07] <CrevsDaak> 200+ mb... 13 mins...
[22:10:28] <lynxlynxlynx> that'll be a painful thing to look over
[22:10:59] <CrevsDaak> i used the sed one liner in doc-player6-bg2-scr-mentions over that
[22:11:15] <lynxlynxlynx> i agree with a more conservative approach, specific tactics for specific problems
[22:11:22] <lynxlynxlynx> too much collateral damage otherwise
[22:12:25] <CrevsDaak> one question
[22:12:44] <CrevsDaak> do you think it's OK to use hacks when using WeiDU instead of normal backup?
[22:13:07] <CrevsDaak> I mean, bacjup doesn't happen when you copy the files outside the game
[22:15:55] <lynxlynxlynx> why copy outside then? the fact that weidu gives you a free backup is welcome
[22:17:56] <lynxlynxlynx> doesn't matter that much during development though
[22:19:56] <CrevsDaak> ok
[22:20:28] <CrevsDaak> actually I say about copying them outside because otherwise you'll have to edit them in the override folder
[22:23:02] <lynxlynxlynx> is that a problem for some reason?
[22:23:29] <lynxlynxlynx> like i said, it doesn't matter for us yet
[22:23:39] <lynxlynxlynx> it's purely a plaintext job for now
[22:26:20] <CrevsDaak> ok
[22:33:02] <lynxlynxlynx> so what's your plan?
[22:37:13] <CrevsDaak> put the sed regexps in to10pp.sh in a perl script and style
[22:37:55] <CrevsDaak> and then check what's going on with uniq on my system
[22:41:43] <lynxlynxlynx> i think it would be better to start afresh
[22:42:13] <CrevsDaak> k, no worries
[22:42:16] <lynxlynxlynx> some of the knowledge can be reused, but even the first 456/654 regexes are too general
[22:42:49] <lynxlynxlynx> if they happen to be in triggers, you could be missing an OR update
[22:43:04] <lynxlynxlynx> so at minimum they need to check indentation level and act accordingly
[22:45:40] <CrevsDaak> actions and OR blocks have the same indentation level
[22:45:58] <CrevsDaak> so we need one for actions and one for triggers
[22:47:19] <lynxlynxlynx> right
[22:47:27] <lynxlynxlynx> positive lookahead it is then
[22:47:35] <CrevsDaak> yeah
[22:47:53] <CrevsDaak> we should check the .ids files to see if they are action or trigger
[22:48:14] <CrevsDaak> unless we're going to first evaluate everything between IF and THEN
[22:48:14] <lynxlynxlynx> interesting approach
[22:48:36] <CrevsDaak> and later everything between RESPONSE #?? and the next response or END
[22:48:41] <lynxlynxlynx> i could swear there are a few naming conflicts
[22:49:00] <lynxlynxlynx> anyway, it's less complicated to do it the dumb way
[22:49:21] <CrevsDaak> I don't remember any naming conflicts
[22:49:36] <lynxlynxlynx> check triggers, fix, mark if we need to copy the whole script block, do the actions bit, resolve
[22:51:59] <-- brada has left IRC (Quit: brada)
[22:52:37] <CrevsDaak> which *.ids should we start using?
[22:52:42] <CrevsDaak> nvm
[22:52:49] <CrevsDaak> the github repo has some
[22:54:04] <CrevsDaak> gaaah, trigger only
[22:56:11] <lynxlynxlynx> i don't think we need to
[22:56:33] <lynxlynxlynx> well, until the compilation, but then you just use the existing one
[22:56:54] <CrevsDaak> I wanted to be sure actions and triggers are all unique
[22:59:42] <lynxlynxlynx> let me just grep that for you
[22:59:57] <lynxlynxlynx> easier to do through our sources than extracting two files per game
[23:02:41] <CrevsDaak> the c++ sources or something else like a default .ids file?
[23:05:55] <lynxlynxlynx> sources
[23:06:04] <lynxlynxlynx> we provide only complementary ids files
[23:06:37] <CrevsDaak> ok
[23:06:42] <CrevsDaak> oh, found them
[23:08:33] <lynxlynxlynx> BitGlobal, ForceMarkedSpell, GlobalAndGlobal, GlobalBAndGlobal, GlobalBitGlobal, GlobalOrGlobal, Help, SetMarkedSpell
[23:08:56] <lynxlynxlynx> a few other candidates
[23:13:20] <CrevsDaak> oh nuts
[23:15:39] --> brada has joined #gemrb
[23:15:57] <lynxlynxlynx> haven't seen the middle ones used before, but the rest are not uncommon (the marked ones are for iwd2)
[23:17:57] <CrevsDaak> I've never played iwd2 so i'm not familiar with it
[23:27:57] <lynxlynxlynx> pretty odd way to script spellcasting if you've only looked at bg
[23:28:14] <lynxlynxlynx> most ai scripts have triggers with side effects
[23:28:44] <lynxlynxlynx> i still haven't made it work completely yet
[23:30:16] <lynxlynxlynx> i'm going to sleep, but won't be going anywhere tomorrow
[23:31:55] <CrevsDaak> ok
[23:32:13] <CrevsDaak> I'm free all day and the whole week until march 9
[23:33:30] --> turtleman_ has joined #gemrb
[23:39:42] <CrevsDaak> Crevs_Away
[23:39:51] --- CrevsDaak is now known as Crevs_Away
[23:43:20] <-- brada has left IRC (Quit: brada)