#gemrb@irc.freenode.net logs for 8 Jan 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:10:25] <Pepelka> [commit] Edheldil: Demo: Correctly convert area's MOS to TIS http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=db39805c392a4ca57ae408864952c25eff9dcf04
[00:13:21] <lynxlynxlynx> wooo
[00:13:27] <edheldil_> :)
[00:13:42] <lynxlynxlynx> so it wasn't my fault either, even better
[00:13:44] <lynxlynxlynx> thanks
[00:14:19] <edheldil_> yeah, looks like dltcep was at fault
[00:14:49] <lynxlynxlynx> what did you use in the end?
[00:14:57] <edheldil_> iesh :)
[00:15:01] <lynxlynxlynx> hehe
[00:15:25] <edheldil_> will have to push the changes
[00:19:24] <lynxlynxlynx> good night
[00:20:16] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[00:38:18] <-- Coriander has left IRC ()
[00:41:37] --> Coriander has joined #gemrb
[01:14:18] --> CamDawg has joined #gemrb
[01:14:52] <edheldil_> they are just aring 'Mazes and Monsters' on a TV here, movie depicting DnD players as cultists, psychos and dead meat
[01:16:26] <CamDawg> Heh, I remember that movie.
[01:17:40] <edheldil_> pushed iesh current state, more or less
[01:23:10] --- ermo is now known as ermo^
[01:23:48] <-- traveler has left IRC (Ping timeout: 245 seconds)
[01:23:48] <-- chiv has left IRC (Ping timeout: 245 seconds)
[01:51:45] <-- edheldil has left IRC (Read error: Connection reset by peer)
[01:55:37] --> brada has joined #gemrb
[02:19:54] <-- edheldil_ has left IRC (Ping timeout: 240 seconds)
[02:28:33] <-- edheldil__ has left IRC (Ping timeout: 276 seconds)
[03:32:36] --> CamDawg1 has joined #gemrb
[03:34:24] <-- CamDawg has left IRC (Ping timeout: 252 seconds)
[03:43:24] --> CamDawg has joined #gemrb
[03:45:31] --> CamDawg2 has joined #gemrb
[03:46:06] <-- CamDawg2 has left #gemrb
[03:46:27] <-- CamDawg1 has left IRC (Ping timeout: 260 seconds)
[03:47:24] --> nutron has joined #gemrb
[03:47:59] <-- CamDawg has left IRC (Ping timeout: 265 seconds)
[04:06:11] --> chiv has joined #gemrb
[04:10:01] <chiv> ive got the hang of github, im beginning to commit the guicommonwindows merges to my repo
[04:11:10] <-- brada has left IRC (Quit: brada)
[05:04:20] --> brada has joined #gemrb
[05:09:49] <brada> chiv: since you are working on guiscripts you should know that the class message window for both BG1 & BG2 is displaying "*" by default
[05:10:08] <brada> dont know when it broke, but it used to work
[05:11:17] <-- WingedHussar has left IRC (Ping timeout: 255 seconds)
[05:11:35] <chiv> well if I find out why, i'll fix it, but I don't think I have been near there yet...
[05:12:04] --> WingedHussar has joined #gemrb
[05:12:33] <brada> sure just something to keep in mind
[05:12:40] <chiv> i was working on a custom guirec but I haven't submitted anything from it
[05:14:29] <chiv> I've thought of even making it like pst, with mouseover hover
[05:31:20] <brada> i dont see why the if (!core->GetAudioDrv()->IsSpeaking()) check in DisplayStringWait is needed at all
[05:31:44] <brada> AFICT bg2 intro forks fine without that check and i dont know wnything else to check
[05:37:15] <brada> by forks or course i mean works...
[05:38:05] <brada> in fact DisplayStringWait doesnt even get called until after the audio is done
[05:38:33] <brada> maybe a later change to scripting negated the need for this?
[05:48:20] <chiv> brada: which class window is broken?
[05:48:32] <brada> when you start a new game
[05:48:38] <brada> and you choose class
[05:49:18] <chiv> so it is, I actually think I noticed that before as well
[05:49:30] <chiv> wierd
[05:54:00] <brada> line 505 of GSUtils right after the string and sound are dispatched there is a SetWait call for the length of the audio
[05:54:28] <brada> so does that negate the need for the isPlaying() check?
[06:08:54] --> WingedHussar1 has joined #gemrb
[06:10:27] <-- DrMcCoy has left IRC (Disconnected by services)
[06:10:39] --> DrMcCoy has joined #gemrb
[06:14:49] <-- WingedHussar has left IRC (*.net *.split)
[06:21:47] <-- brada has left IRC (Quit: brada)
[07:05:04] --> edheldil__ has joined #gemrb
[07:22:37] --> miha has joined #gemrb
[07:24:27] <-- nutron has left IRC (Remote host closed the connection)
[07:33:45] <chiv> https://github.com/chilvence/gemrb/commits/master <-- initial success with the merging. I hope my commits aren't too big for you still, feel free to bitchslap me until I get the point. If I was not in full understanding of how anything in there works, I have been bluntly honest about it in the log.
[07:33:47] <Pepelka> Commit History · chilvence/gemrb · GitHub
[07:38:43] <-- edheldil__ has left IRC (Ping timeout: 246 seconds)
[08:01:01] <-- WingedHussar1 has left IRC (Quit: WingedHussar1)
[08:14:10] --> WingedHussar has joined #gemrb
[08:46:14] --> edheldil has joined #gemrb
[08:46:15] --- ChanServ gives channel operator status to edheldil
[09:38:48] --> lynxlynxlynx has joined #gemrb
[09:38:48] <-- lynxlynxlynx has left IRC (Changing host)
[09:38:48] --> lynxlynxlynx has joined #gemrb
[09:38:48] --- ChanServ gives channel operator status to lynxlynxlynx
[10:08:34] <edheldil> chiv: I just can't see an advantage of it ... and it reminds me of Perl too much :)
[10:43:03] --> traveler has joined #gemrb
[12:13:00] <lynxlynxlynx> chiv: import GUIClasses ##this may not be absolutely necessary? <-- well pst needs it, so you can put it in the other import ifdef
[12:15:18] <lynxlynxlynx> or just leave both, there's not much damage in importing them
[14:18:31] --> avenger has joined #gemrb
[14:19:22] <avenger> yeah, wjp is right, i thought that patch changed LoadWindow :)
[14:21:15] <-- lynxlynxlynx has left IRC (Ping timeout: 276 seconds)
[14:24:34] * avenger gates lynxlynxlynx.
[14:27:21] <-- Pepelka has left IRC (Remote host closed the connection)
[14:27:47] --> Seniorita has joined #gemrb
[14:27:49] <edheldil> avenger: I think dltcep generates wrong tis files from mos
[14:28:28] <-- Seniorita has left IRC (Remote host closed the connection)
[14:28:45] --> Seniorita has joined #gemrb
[14:33:23] <avenger> current dltcep or several versions older?
[14:33:36] <avenger> current one might be affected by bgee hacks
[14:34:58] <avenger> btw, why would you want to generate tis from mos O_O
[14:36:26] <avenger> mos is the minimap format, it works the other way: bmp to tis or mos, tis to bmp or mos. mos isn't too useful to convert back, because it is downscaled
[14:44:27] <edheldil> oh, possibly it was from bmp - the issues lynx had with striped area were caused by dltcep creating tis like that
[14:45:11] <edheldil> I used some older version of dltcep, I think
[14:47:03] --> kida has joined #gemrb
[14:50:34] <-- avenger has left IRC (Quit: Page closed)
[14:58:20] <-- miha has left IRC (Quit: Lost terminal)
[15:04:30] --> lynxlynxlynx has joined #gemrb
[15:04:30] <-- lynxlynxlynx has left IRC (Changing host)
[15:04:30] --> lynxlynxlynx has joined #gemrb
[15:04:30] --- ChanServ gives channel operator status to lynxlynxlynx
[15:21:05] <-- Cable__ has left IRC (Ping timeout: 265 seconds)
[15:23:45] <lynxlynxlynx> eh, a near miss
[15:29:18] --> CamDawg has joined #gemrb
[15:33:23] <-- WingedHussar has left IRC (Ping timeout: 248 seconds)
[15:33:46] --> WingedHussar has joined #gemrb
[15:34:57] --> Cable__ has joined #gemrb
[15:51:10] --> miha has joined #gemrb
[16:12:20] <-- WingedHussar has left IRC (Quit: WingedHussar)
[16:12:39] --> brada has joined #gemrb
[16:18:13] <brada> chiv: compare view is your friend
[16:18:13] <brada> https://github.com/chilvence/gemrb/compare/gemrb:master...master
[16:18:18] <Seniorita> Comparing gemrb:master...chilvence:master · chilvence/gemrb · GitHub
[16:18:55] <brada> at first i was irked cuz you didnt do a branch, but now i know you can compare across forks :)
[16:25:01] <brada> lynx: did you see what i said about DisplayStringWait?
[16:25:15] <lynxlynxlynx> cutscenes
[16:25:46] <lynxlynxlynx> the bg2 cutscene when you get out of the first dungeon is a good test
[16:30:25] <brada> testing that now
[16:31:17] <chiv> brada: cheers, I was looking for that - is there a side by side compare do you know?
[16:34:45] <brada> ok i see the problem now
[16:36:28] <brada> chiv: i dont know about side by side
[16:36:47] <-- Dakkus has left #gemrb
[16:37:35] <chiv> edheldil: I would answer, but I am not really sure what you are referring to by ' it ' ;)
[16:40:49] <chiv> oh well, its still pretty need to be able to compare entire projects...
[16:45:35] <chiv> lynxlynxlynx: I was just not sure why pst imports guiclasses there, when none of the other games do
[16:45:48] <lynxlynxlynx> search the file :)
[16:49:42] <chiv> yep, I see i t now
[16:53:08] <-- kida has left IRC (Remote host closed the connection)
[16:55:58] <chiv> by the way, whats the method for running the demo game? I tried the other week but just got stuck with startup errors so I moved on...
[16:56:24] --> kida has joined #gemrb
[17:01:35] <chiv> I saw someone doing some crazy nice art on gibberlings 3 before it started going to hell
[17:05:41] <chiv> http://forums.gibberlings3.net/index.php?showtopic=22118&st=165
[17:05:42] <Seniorita> Other cities than Athkatla and BG... - The Gibberlings Three Forums - Page 12
[17:09:18] <brada> chiv: you need ttf plugin for starters
[17:11:15] <chiv> I should get that...
[17:12:24] <chiv> to be honest I installed the gemrb dependencies years ago and haven't looked at them since
[17:15:15] <lynxlynxlynx> those area bgs are unusable though
[17:15:53] <chiv> how come?
[17:20:37] <Seniorita> [commit] Brad Allred: Actions: use the wait time instead of relying on wether the audio driver "IsSpeaking" http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=6480749f0d72c546808eb58f76488bc2df1560e8
[17:20:38] <Seniorita> [commit] Brad Allred: Actions: dont need to know about audio http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=6b4ea01d02f2f6862a6b39217d3ace877201db38
[17:20:39] <Seniorita> [commit] Brad Allred: remove IsSpeaking() from Audio http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=815907861f4dd024780bd049d46a455abef1752f
[17:20:44] <lynxlynxlynx> reused artwork
[17:20:46] <brada> you also need png importer
[17:20:52] <brada> ^ for demo
[17:20:55] <lynxlynxlynx> tainted license basically
[17:21:11] <chiv> doh
[17:21:25] <lynxlynxlynx> https://github.com/chilvence/gemrb/commit/37c2154fa7888d5db45cf7b5e33132df61578cb9 <-- this one should be split and added to the relevant commits
[17:21:27] <Seniorita> Fix error merging follow button toggler · 37c2154 · chilvence/gemrb · GitHub
[17:22:54] <chiv> yeah, i wasn't sure how to fix it after i had pushed...
[17:23:05] <chiv> good think i have my own sandbox eh
[17:23:11] <chiv> *thiing
[17:23:30] <chiv> this damn keyboard... its too small
[17:24:59] <lynxlynxlynx> you'll have some fun heh
[17:29:10] <lynxlynxlynx> brada: GameTime is a ieDword
[17:29:17] <brada> doh
[17:29:35] <brada> i have to run to work. feel free to fix it
[17:30:05] <-- brada has left IRC (Quit: brada)
[17:32:40] <lynxlynxlynx> - SetPSTGamedaysAndHourToken ()
[17:32:40] <lynxlynxlynx> + GUICommon.SetGamedaysAndHourToken ()
[17:32:45] <lynxlynxlynx> that is also fishy
[17:33:13] <lynxlynxlynx> golden rule, one thing per commit
[17:33:19] <chiv> well, it doesnt seem to work as it is is anyway, so I had that pinned for investigation
[17:33:46] <chiv> I can do the whole commit thing again, if you would prefer
[17:34:09] <lynxlynxlynx> they are different due to different tokens iirc
[17:34:24] <lynxlynxlynx> adding extra bugs can't help in any way
[17:34:52] <lynxlynxlynx> so far you only need to fix a few
[17:38:20] <lynxlynxlynx> for the previous one, the end of https://github.com/chilvence/gemrb/commit/3ee481db4753848ef7cd2ad6aa5b3e350f4058a0 needs changing too
[17:38:22] <Seniorita> Merging a selection of nebulous items together, to see what explodes · 3ee481d · chilvence/gemrb · GitHub
[17:40:31] <chiv> seperate commits?
[17:40:39] <lynxlynxlynx> the two functions you comment out at the end of it can be just removed too, you handled that in earlier commits
[17:41:04] <chiv> looking back at the clock, I need to do something about it
[17:41:27] <chiv> it doesn't seem to work in vanilla gemrb
[17:45:50] <lynxlynxlynx> https://github.com/chilvence/gemrb/commit/02eb3ec2a5c962588dc0813a815f3d5d9a511330 <-- second chunk here looks bad too, the current version is better
[17:45:52] <Seniorita> Blindly synching the selection changed handler. · 02eb3ec · chilvence/gemrb · GitHub
[17:46:01] <lynxlynxlynx> if it doesn't work for pst, ifdef it
[17:50:28] <chiv> you mean @1376?
[17:51:41] <lynxlynxlynx> 1179, i don't see 1376 anywhere
[17:51:57] <chiv> ah I thought you meant the second file...
[17:53:34] <lynxlynxlynx> and that clock foo will cause you to need updating https://github.com/chilvence/gemrb/commit/c1684470a508a22c0d1bda6fa867ea9cbc3b7169 too
[17:53:36] <Seniorita> That synching feeling again · c168447 · chilvence/gemrb · GitHub
[17:58:05] <lynxlynxlynx> https://github.com/chilvence/gemrb/commit/07635d85778a6d0be683dacd93aaa85093d7b1f6
[17:58:05] <lynxlynxlynx> you set Label conditionally now, so any users would still break if it is None
[17:58:08] <Seniorita> Merge portrait frame/hp · 07635d8 · chilvence/gemrb · GitHub
[17:59:42] <lynxlynxlynx> so it either makes no sense or needs extra checking on use
[17:59:57] <chiv> doh.. that was a bit stupid
[18:00:57] <chiv> but then iirc I could not even find where it was used
[18:01:19] --> Yoshimo has joined #gemrb
[18:02:33] <chiv> thanks for the review by the way, its making it much easier to weed out the problems
[18:02:36] <lynxlynxlynx> https://github.com/chilvence/gemrb/commit/eb21578115d9f5ce3ca225ab0babe4a060753075
[18:02:42] <Seniorita> Primary Objective Achieved · eb21578 · chilvence/gemrb · GitHub
[18:03:06] <lynxlynxlynx> ButtonHP is now set in a local scope, you should get UnboundLocalSomething exceptions
[18:04:10] <lynxlynxlynx> instead of indenting both portrait paths, i would prefer if you moved each to a function in a separate commit and then use it here
[18:04:32] <lynxlynxlynx> that's it
[18:04:46] <lynxlynxlynx> quite ok for a first ever try
[18:05:07] <chiv> awesome, much less trouble than I expected to be honest
[18:05:13] <lynxlynxlynx> now for some political practice err i mean history rewriting
[18:05:23] --> Avenger has joined #gemrb
[18:05:23] --- ChanServ gives channel operator status to Avenger
[18:05:29] <chiv> heh
[18:05:30] <lynxlynxlynx> ojla
[18:05:40] <Avenger> hi everyoe
[18:05:45] <Avenger> +n :)
[18:06:03] <chiv> i have added all those comments in the compare, so i will get back on that but now I need to eat
[18:06:47] <lynxlynxlynx> Avenger: what i pinged you about is iwd2 cre spell indices
[18:07:08] <Avenger> domain spells?
[18:07:24] <lynxlynxlynx> they are not problematic for wizard or cleric spells, but the other spellbooks get odd values often
[18:07:29] <lynxlynxlynx> like domain spells, yes
[18:07:31] <Avenger> i can look at iwd2 if you have a question
[18:07:47] <Avenger> the indices are not fitting into the domain list?
[18:07:52] <lynxlynxlynx> i made them work, but i had to repeat the table lookup, since the indices are off
[18:07:56] <lynxlynxlynx> yeah
[18:08:15] <lynxlynxlynx> dltcep doesn't display much, just confirms that it can also read the count
[18:09:00] <lynxlynxlynx> http://sprunge.us/PBgh?diff <-- this is the hack
[18:09:29] <lynxlynxlynx> i've tried ANDing the index with the kit, but no luck on the two samples
[18:10:07] <lynxlynxlynx> eg, starting with a predefined party (so the char still has domain spells), we want to give a level 9 spell to a cleric :)
[18:12:07] <-- kida has left IRC (Ping timeout: 260 seconds)
[18:12:33] <lynxlynxlynx> ”Unresolved spell index: 300 level:7, type: 1” or “index: 253 level:1, type: 6” <-- from the todo, so apparently we also go bad on some cleric spells
[18:13:12] <lynxlynxlynx> level1 cleric and druid spells seem to be loaded fine now though
[18:13:59] <lynxlynxlynx> didn't check if there's any exclusion going on though
[18:15:03] <lynxlynxlynx> bbl
[18:17:41] <Avenger> yes, i can see that dltcep cannot resolve the domain spells either
[18:17:50] <Avenger> first i will have to fix that
[18:17:57] <Avenger> once i fixed that, i'll surely have a working model
[18:18:13] <Avenger> but i have some clock punching time with bgee too :)
[18:21:29] --> brada has joined #gemrb
[18:27:49] <Seniorita> [commit] Brad Allred: Actions: time is a dword http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=a0e923b9436fc2075d490bfb35d70bfa94bbf4e1
[18:42:27] <-- Yoshimo has left IRC (Quit: Yoshimo)
[18:56:48] <lynxlynxlynx> dltcep would be fine just displaying the index
[18:57:04] <brada> lynx: AmbientMgrAL has todo for removing openal dependancies, but i dont see any
[18:57:22] <brada> and removing the openal headers doesnt break compiling
[18:57:24] <lynxlynxlynx> except for the name?
[18:57:43] <lynxlynxlynx> it was a task long ago, but it's quite possible zefklop and wjp fixed it completely
[18:58:02] <lynxlynxlynx> AmbientMgrAL shouldn't have that suffix if it is backend agnostic though
[18:58:16] <brada> sure
[18:58:33] <brada> todo says to move it
[18:58:37] <brada> dont know where
[19:02:53] <fuzzie> into core
[19:03:19] <fuzzie> right now it depends on SDL everywhere though
[19:04:44] <brada> yes it still does have that dep
[19:04:53] <brada> but no openal dep
[19:04:59] <Seniorita> [commit] Brad Allred: remove useless #include "Audio.h" lines http://gemrb.git.sourceforge.net/git/gitweb.cgi?p=gemrb/gemrb;a=commitdiff;h=e4cbfba396fa0fe100b2fdd91447ac479bf9d461
[19:05:12] <brada> and that huge block of openal header includes is useless
[19:05:18] <brada> at least on my platform
[19:06:20] <fuzzie> but SDLAudio doesn't support ambients right now due to lack of this thing
[19:07:51] <brada> what are you getting at fuzzie?
[19:09:52] <fuzzie> well, it would be good to move it :P
[19:11:21] <fuzzie> which is to say, it isn't just a useless todo
[19:12:59] <brada> i wasnt saying it ws useless
[19:13:11] <brada> i was saying i dont see a dep for openal and we should finish that todo
[19:13:58] <fuzzie> sure, I'm just being positive about it :)
[19:14:19] <fuzzie> or trying to be and failing, apparently :P
[19:15:47] <fuzzie> It would be nice to merge it right into AmbientMgr.cpp but the threading is icky.
[19:15:54] <brada> yes
[19:16:45] <fuzzie> I don't see at a glance why we can't just update it every frame from the main loop or similar.
[19:17:47] <fuzzie> The ambient intervals are in seconds, so it doesn't exactly need real-time or anything..
[19:17:51] <fuzzie> wjp: Do you remember this?
[19:20:21] <Seniorita> [wiki] developers:ietesting http://www.gemrb.org/wiki/doku.php?id=developers:ietesting&rev=1357672806&do=diff
[19:26:24] <Seniorita> [wiki] developers:ietesting http://www.gemrb.org/wiki/doku.php?id=developers:ietesting&rev=1357672836&do=diff
[19:26:48] <wjp> hm?
[19:27:56] <wjp> hm, no, don't remember
[19:30:45] --> edheldil_ has joined #gemrb
[19:39:21] <wjp> it does seem a bit overly complicated for what it does
[19:44:36] <Seniorita> [wiki] contribute - split intwo: code/nocode http://www.gemrb.org/wiki/doku.php?id=contribute&rev=1357674196&do=diff
[19:45:23] <chiv> hey, does the wiki have a personal user page that I can edit?
[19:48:01] <lynxlynxlynx> create one
[19:48:28] <lynxlynxlynx> i suggest users:chiv in case it becomes a trend
[19:48:37] <chiv> ok. good idea
[19:52:50] <-- edheldil_ has left IRC (Ping timeout: 255 seconds)
[20:04:00] --> edheldil_ has joined #gemrb
[20:07:16] <lynxlynxlynx> fuzzie: do you know about any large, odd-shaped travel regions that are not on map borders?
[20:08:57] <lynxlynxlynx> actually, i can't think of many that aren't rectangular themselves, so i'm thinking about just committing this http://sprunge.us/WWaD?diff
[20:38:29] <fuzzie> um
[20:42:27] <fuzzie> well I guess it doesn't matter for now
[20:43:04] <fuzzie> this isn't at all how it's handled in original
[20:43:30] <fuzzie> original handles it on click, by queuing a MoveToPoint then a LeaveAreaName (both on the group)
[20:43:53] <-- miha has left IRC (Ping timeout: 252 seconds)
[20:43:54] <fuzzie> so all the ST_TRAVEL hacks in that function should die anyway
[20:46:03] --> miha has joined #gemrb
[20:52:59] <-- edheldil_ has left IRC (Ping timeout: 252 seconds)
[20:54:12] <-- miha has left IRC (Ping timeout: 276 seconds)
[20:55:15] --> miha has joined #gemrb
[21:00:35] <lynxlynxlynx> i'll commit it for now then and let you remove it, hopefully soon ;)
[21:01:23] <Seniorita> [wiki] users:chiv - created: chiv's fridge list http://www.gemrb.org/wiki/doku.php?id=users:chiv&rev=1357678770&do=diff
[21:01:40] <lynxlynxlynx> http://paste.debian.net/222661/ <-- this is the data; the launch point is about 100 units away from the bbox, but on the unreachable side
[21:01:42] <Seniorita> debian Pastezone
[21:02:15] <lynxlynxlynx> so hopefully leaveareaname gives up in the end and ignores range
[21:03:50] <fuzzie> yes, it does the right thing, I just forget how
[21:04:57] <fuzzie> LeaveAreaName is heaps of fun because pst has a heap of hard-coded logic in it. :)
[21:06:47] <fuzzie> unfortunately irc logs don't seem to have anything useful except that.
[21:06:51] <|Blaze|> do you know that BG2 has 3 copies of the area transition code in it? Area transitions via clicking on the world map, area transitions via script, and area transitions via clicking on a door where the whole party needs to make the transition
[21:06:51] <lynxlynxlynx> i'd start with the bg2 or how or iwd2 way it works, it is likely better in an ergonomic sense
[21:07:10] <lynxlynxlynx> heh
[21:07:13] <|Blaze|> it's such a mess
[21:08:14] <fuzzie> if you think bg2 is a mess, be glad you didn't look at pst ;P
[21:08:36] <fuzzie> and I always check bg2 first because we have symbols for it and I am lazy.
[21:08:57] <|Blaze|> I can only imagine. That was branched before BG1 shipped I think
[21:09:03] <fuzzie> yep
[21:09:27] <lynxlynxlynx> what i was trying to say is that hopefully we don't need to reproduce all the hacks from pst
[21:09:40] <fuzzie> lynxlynxlynx: well, we do :-) but I think we already started on some of them elsewhere
[21:09:46] <Avenger> lynx the domain spells use listspl to resolve number to resref
[21:09:58] <lynxlynxlynx> hah
[21:10:09] <Avenger> just found it now :D
[21:10:13] <chiv> when is leaveareaname used?
[21:10:26] <Avenger> hey |Blaze| :)
[21:10:30] <|Blaze|> hey
[21:10:30] <lynxlynxlynx> thanks
[21:10:35] <fuzzie> chiv: it's queued when you click on a travel region
[21:10:58] <lynxlynxlynx> Avenger: i'm pretty sure some others use it then too, domain spells weren't the only ones with bad indices
[21:11:20] --> edheldil_ has joined #gemrb
[21:11:26] <Avenger> probably everything uses listspl for number->resref
[21:11:30] <Avenger> that is actually a good news
[21:11:41] <chiv> so is it for advancing the gametime or something?
[21:12:02] <lynxlynxlynx> leaving areas? :)
[21:12:26] <chiv> haha
[21:12:28] <fuzzie> chiv: yes, as the name suggests, it handles leaving an area :P
[21:13:16] <lynxlynxlynx> great, lisspl has more than 300 entries - i don't have any report of higher bad indices
[21:14:10] <lynxlynxlynx> yep, matches this cleric of talos too
[21:14:17] <lynxlynxlynx> weee
[21:14:30] <chiv> I only asked, because on thing I noticed with pst is that when you are at the start, the travel is quite basic, but when you get the worldmap, travelling just to the adjacent area adds hours and hours...
[21:14:58] <fuzzie> ah, well, yes, different code :)
[21:15:01] <lynxlynxlynx> travel time is stored
[21:15:48] <chiv> i wondered if it was doing some voodoo to actually iterate over the areas you run past
[21:16:00] <fuzzie> alas, no
[21:16:12] <fuzzie> there's a worldmap file which has links between the areas you can worldmap travel between
[21:16:12] <chiv> oh well, just a thought
[21:16:21] <fuzzie> and it'll step through the necessary entries in that file
[21:17:38] <lynxlynxlynx> yeah, adjacency is not as it seems
[21:17:44] <chiv> I have to wonder what it would be doing though, because I grit my teeth and managed to step through 90% of the game with GRB
[21:17:55] <-- miha has left IRC (Quit: leaving)
[21:18:34] <fuzzie> GRB? pst is a bit disorganised in that sense though
[21:18:53] <chiv> gemrb
[21:18:59] <fuzzie> oh right :)
[21:19:16] <fuzzie> you didn't notice how painful it is to leave areas in gemrb sometimes? :P
[21:19:25] <chiv> actually... no
[21:19:29] <chiv> heh
[21:19:34] <Avenger> chiv: baator has a reliable crasher though?
[21:19:43] <fuzzie> did you play with a party?
[21:19:50] <chiv> yeah, it was the one area i skipped
[21:19:52] <chiv> yes
[21:19:53] <lynxlynxlynx> Avenger: and two projectiles
[21:20:13] <Avenger> any pointers on what crashes? like a crashdump?
[21:20:18] <lynxlynxlynx> all beatufully documented
[21:20:31] <lynxlynxlynx> the projectiles are just missing files on our side
[21:20:32] <chiv> did most of the dialogue with dakkon, dont really know the other characters thoroughly
[21:20:46] <lynxlynxlynx> adder.spl and a pro for ball lightning
[21:20:53] <Avenger> well...
[21:20:59] <Avenger> those two are very complex
[21:21:09] <chiv> dummys would be an advantage
[21:21:17] <lynxlynxlynx> sure or you wouldn't have left them out, but we should crash
[21:21:19] <fuzzie> can't we fix it to at least silently fail though?
[21:21:24] <lynxlynxlynx> exactly
[21:21:26] <chiv> yes
[21:21:29] <Avenger> the adder is a snake that goes in spiral and sine wave path
[21:21:35] <lynxlynxlynx> +not somewhere
[21:21:43] <lynxlynxlynx> hah
[21:21:54] <chiv> its irritating in the game anyway
[21:22:03] <chiv> takes half an hour to actually hit
[21:22:40] <chiv> ball lightning is more fun, lightning fx are always a winner
[21:26:19] <chiv> I think particle fx can come later though... at the moment its just impossible to run any map that has any kind of spellcaster
[21:28:00] <lynxlynxlynx> don't remember what it does, but i suggest we just use bg2's chain lightning for now
[21:28:33] <lynxlynxlynx> adder is only missing a spell it seems, so a direct damage dummy sounds good enough
[21:28:33] <Avenger> so where is the baator bug? i doubt it is the projectiles
[21:28:40] <chiv> afaik the particle fx are just visual, the ball lightning ... balls.. are just like another magic missile
[21:29:03] <chiv> Avenger: as i said, its the only area I skipped past
[21:29:31] <chiv> but it crashes as soon as you warp in from the outlands for me
[21:30:12] <chiv> i still need to make another debug built and analyse those types of things
[21:30:21] <chiv> *build
[21:35:43] <chiv> my primary motivation is I kept seeing people ask about how well pst works, but there was not really a definitive analisys
[21:36:57] <chiv> so now the answer is 'alot better than you probably imagined' :)
[21:37:56] --> nutron has joined #gemrb
[21:40:10] <chiv> I do wonder why BI dumped alot of the cooler things from the PST branch though...
[21:45:01] <fuzzie> it was all a horrible hack which could never be rescued
[21:46:26] <chiv> do you enjoy reverse engineering it though or is it more like you are glad when it is over? :)
[21:49:59] <fuzzie> I generally just leave it to Avenger. ;)
[21:51:19] <Avenger> first it seemed scary, but after five or so years, it is kinda familiar.
[21:52:35] <chiv> I think what you guys do is nothing short of magic... I have no idea how you can pull sense from some giant binary file
[21:52:51] <Avenger> chiv: did you get the baator movie?
[21:53:04] <chiv> sorry, cant remember...
[21:53:05] <Avenger> ah ok, after movie: crash
[21:53:36] <Avenger> some script
[21:53:47] <chiv> it was more of a scouting run, so details are very thin on the page
[21:54:05] <Avenger> ar1000 area script itself crashes
[21:54:21] <fuzzie> any idea why? :)
[21:55:01] <Avenger> yes
[21:55:05] <Avenger> ChangeAIScript("ar1000ch",DEFAULT) then continue
[21:55:11] <Avenger> HUUH
[21:56:04] <Avenger> the scripters of pst were definitely crazy people :D
[21:56:34] <Avenger> but it is a good test for sure
[21:56:55] <lynxlynxlynx> test of resilience to insanity
[21:57:29] <chiv> so... it changes to another script... then tries to continue itself?
[21:57:46] <Avenger> yes basically it kills itself while still trying to do something
[21:57:52] <fuzzie> ah
[21:57:54] <fuzzie> that is a known bug
[21:58:01] <fuzzie> it happens somewhere else too
[21:58:18] <chiv> just making sure, I wasn't sure if my stupid was showing itself...
[21:58:18] <fuzzie> we don't keep a reference to scripts while they run and we just delete them..
[21:58:42] <Avenger> originally i intended to do that
[21:59:02] <fuzzie> well it boils down to my fault anyway :p
[21:59:24] <fuzzie> but I thought someone fixed it with a hack
[21:59:29] <Avenger> can you fix it?
[22:00:22] <fuzzie> oh, it was you, and you added a hack elsewhere
[22:00:52] <Avenger> hmm, well, this is just a guess of the crash location
[22:01:29] <Avenger> but it seems the case. The canary variable is totally garbage
[22:01:48] <fuzzie> oh, not 0xdddddddd?
[22:02:02] <fuzzie> brad had this but he managed to nicely get 0xdddddddd out of it
[22:02:56] <fuzzie> anyway, it's almost certianly that action. because, it triggers the stupid bug to do with us executing actions in the wrong place though, so it's not trivial to fix.
[22:03:37] <Avenger> cdcdcdcd
[22:03:46] <Avenger> at least some other variables near it
[22:04:03] <Avenger> the canary itself is garbage
[22:04:06] <Avenger> and i'm on windows
[22:04:20] <Avenger> cd is the deleted area, right?
[22:05:00] <-- brada has left IRC (Quit: brada)
[22:05:20] <fuzzie> probably best just to add it to the List Of Things Which Must Be Fixed for now
[22:05:29] <fuzzie> and maybe see if you can make the projectiles not crash
[22:08:04] --> brada has joined #gemrb
[22:10:35] <brada> is it jsut PST that this happens?
[22:11:06] <brada> i can tell you another place it happens if you need
[22:12:55] --> CamDawg1 has joined #gemrb
[22:13:51] <-- CamDawg has left IRC (Ping timeout: 265 seconds)
[22:17:30] --> CamDawg has joined #gemrb
[22:17:37] <Avenger> brada i think it happens only if there is a continue after changeaiscript killing its own script. But otherwise, it is not pst specific
[22:17:43] <-- CamDawg1 has left IRC (Ping timeout: 265 seconds)
[22:18:07] <Avenger> i don't know what would be the expected behavior
[22:18:46] <Avenger> if the rest of the script (after continue) shouldn't run, then the fix is easy
[22:22:08] <-- CamDawg has left IRC (Ping timeout: 255 seconds)
[22:22:10] <chiv> one would imagine that 'changeaiscript' is not meant to leave the original script running...
[22:22:27] <chiv> but does that just show how little I know IE?
[22:22:41] <lynxlynxlynx> scripting assumptions are dangerous
[22:25:00] <Avenger> even changing it to clearactions proved to be wrong
[22:25:19] <Avenger> it should definitely execute the already queued actions
[22:25:28] <Avenger> no idea about the stuff after continue
[22:25:57] <Avenger> probably we do this wrong, we should probably queue them in the same run
[22:27:53] <-- Avenger has left IRC (Quit: bye!)
[22:30:58] <brada> well certainly it wont break things more then they already are to just ignore the rest of the script
[22:31:09] --> edheldil__ has joined #gemrb
[22:31:16] <brada> even if the result is wrong at least it doesnt crash
[22:31:22] <chiv> I can imagine this sort of scenario: Area designer makes bad script. Game programmer gets shakedown next morning about broken scripting and told to fix it. Game programmer bashes head against desk in despair and hacks code to avoid stupid input.
[22:42:50] <brada> is say do this "Avenger: if the rest of the script (after continue) shouldn't run, then the fix is easy"
[22:44:39] <lynxlynxlynx> he already tried if i understand correctly
[22:48:05] <brada> well its either that or annoy fuzzie into submission :p
[22:48:08] <-- brada has left IRC (Quit: brada)
[23:18:16] <lynxlynxlynx> bah, evil deities, always noncompliant
[23:18:57] <lynxlynxlynx> there are some 10 knights of xvim, yet none have the same scripting name as we print
[23:19:22] <lynxlynxlynx> all have the same levels, 18 pally, 1 cleric
[23:19:33] <lynxlynxlynx> from the ones i've sampled, none have any casting levels
[23:20:11] <lynxlynxlynx> hmm, maybe it's the wisdom bonus