#gemrb@irc.freenode.net logs for 2 Mar 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:32:43] <-- Bo_Thomsen1 has left IRC (Quit: Leaving.)
[02:20:48] <-- edheldil_ has left IRC (Ping timeout: 264 seconds)
[05:44:32] --> raevol has joined #GemRb
[07:10:49] --> Bo_Thomsen1 has joined #GemRb
[07:36:13] --> lynxlynxlynx has joined #GemRb
[07:36:13] <-- lynxlynxlynx has left IRC (Changing host)
[07:36:13] --> lynxlynxlynx has joined #GemRb
[07:36:13] --- ChanServ gives channel operator status to lynxlynxlynx
[07:53:44] <-- raevol has left IRC (Quit: Leaving.)
[07:54:46] <-- Bo_Thomsen1 has left IRC (Quit: Leaving.)
[08:04:21] --> lubos has joined #GemRb
[08:15:15] --> edheldil_ has joined #GemRb
[08:20:04] --> adominguez has joined #GemRb
[08:21:38] <-- adominguez has left IRC (Remote host closed the connection)
[08:26:13] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[08:40:26] <lubos> Hi all, I found next source files without licence headers (plugin BMPWriter)
[08:51:21] <lubos> next bg1/CharGenGui.py
[08:58:00] <lubos> core/System/snprintf.cpp is suspected
[09:00:54] <lubos> includes/operatorbool.h
[09:08:41] --> adominguez has joined #GemRb
[09:41:39] <edheldil> btw, if anyone gets to fixing it, consider changing the date on Autodetect scripts to 2011
[09:51:17] <lubos> next in GUIScripts: BGCommon.py, ie_spells.py, include.py
[09:58:43] <-- |Cable| has left IRC (Remote host closed the connection)
[10:31:39] <edheldil> btw. if it is of any help to you, there is admin/check_copyrights.sh or some such
[10:36:33] <lubos> thx. Debian has licensecheck (package devscripts)
[12:21:26] <lynxlynxlynx> cool
[12:22:15] <lynxlynxlynx> i'll get a few android phones from on of our sponsors, meaning i can also try that gemrb build :)
[12:40:05] <pupnik> how the hell can android run gemrb
[12:40:10] <pupnik> it's missing *everything*
[12:40:28] <edheldil> hehe
[13:12:04] <lynxlynxlynx> not really, plus gemrb isn't openoffice
[13:13:27] <pupnik> all the libraries built for gemrb, are they available to other android apps?
[13:13:37] <pupnik> is there a repository?
[13:13:40] <pupnik> no, no
[13:21:29] <lynxlynxlynx> no idea
[13:22:05] <lynxlynxlynx> probably better in cyanogenmod
[13:47:44] <lynxlynxlynx> fuzzie: dlg/bg2/aerie.d block 2
[13:48:00] <lynxlynxlynx> any quick guesses why it doesn't run fully?
[13:48:41] <lynxlynxlynx> SetGlobal did run and so did probably the last StartDialogueNoSet,
[13:48:52] <lynxlynxlynx> but not the stuff in between
[13:56:45] <-- adominguez has left IRC (Remote host closed the connection)
[14:07:41] <pupnik> see but android has no ecosystem for this
[14:08:04] <pupnik> no repository, no community around building libs - just individuals uploading stuff to mediafire
[14:11:53] <edheldil> android supports 'homo homini lupus' cooperation model, just like AppStore ;-)
[14:39:14] --> adominguez has joined #GemRb
[14:40:40] <fuzzie> well, google rather sabotaged the whole repos thing with the 'Android Market' which isn't on half the android devices
[14:41:59] <fuzzie> lynxlynxlynx: we don't wrap dialog actions with a NoInterrupt at begin/end, i assume it is interrupted
[14:47:07] <lynxlynxlynx> should i try with that?
[14:47:36] <fuzzie> sure
[14:48:07] <fuzzie> wait, is that the last piece in a dialog?
[14:49:03] <fuzzie> i guess it is. so yes, i would try with that. i can't see what on earth would trigger another dialog though, the stuff in the middle will always run if you get to StartDialogueNoSet
[15:00:32] <lynxlynxlynx> the save is after-the-fact, so i'm not sure - maybe he started another dialog manually
[15:01:21] <lynxlynxlynx> will wait a bit if he has a good one, otherwise i'll have to make my own (boohoo, more playing games :])
[15:03:25] <fuzzie> so is normal gemrb broken about shooting through walls?
[15:03:57] <fuzzie> i fixed that for the first dungeon area last year, but before all these searchmap changes
[15:12:38] <edheldil> lynxlynxlynx: if you need, I can send you HoW saves
[15:12:54] <edheldil> or rather put them on my web
[15:13:59] <edheldil> (it's actually in the IWD part, since it's still only in Valley of Shadows)
[16:03:19] --> pupnik_ has joined #GemRb
[16:06:35] <-- pupnik has left IRC (Ping timeout: 240 seconds)
[16:33:03] --> Maighstir has joined #GemRb
[16:53:04] <-- lubos has left IRC (Quit: Leaving.)
[16:55:01] <-- Maighstir has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[16:55:36] --> Maighstir has joined #GemRb
[17:06:31] --> Bo_Thomsen has joined #GemRb
[17:13:38] <-- xrogaan has left IRC (Ping timeout: 260 seconds)
[17:40:51] <lynxlynxlynx> edheldil: thanks, but i have my own from the gemrb playthroughs
[17:41:10] <lynxlynxlynx> fuzzie: yeah, definitely something to check before the release
[17:41:35] <lynxlynxlynx> i haven't experienced it, but i haven't playtested much after those changes
[17:44:11] --- pupnik_ is now known as pupnik
[17:51:37] --> xrogaan has joined #GemRb
[17:51:37] <-- xrogaan has left IRC (Changing host)
[17:51:37] --> xrogaan has joined #GemRb
[17:52:31] <fuzzie> hmm
[18:01:10] <fuzzie> well, i stared at the code a bit and i don't see issues
[18:03:42] <fuzzie> we do use floating-point in Map::IsVisible.
[18:03:50] <fuzzie> but i guess broken fp would show up in more visible ways.
[18:04:27] <-- adominguez has left IRC (Quit: Saliendo)
[18:04:34] <fuzzie> but no nice obvious things like the use of 'char'.
[18:04:48] --> barra_home has joined #GemRb
[18:28:44] <-- barra_home has left IRC (Quit: Verlassend)
[18:28:50] --> |Cable| has joined #GemRb
[19:17:05] <xrogaan> sup
[19:17:15] <xrogaan> do gemrb require python to run ?
[19:17:31] <xrogaan> or just the libraries to build gemrb ?
[19:19:51] <fuzzie> you don't need the python interpreter, if that's what you mean
[19:19:57] <fuzzie> but you do need the python libraries at runtime
[19:25:24] <xrogaan> thanks
[19:26:18] <xrogaan> is python 2.7 okay ?
[19:27:45] <fuzzie> yes
[19:28:05] <xrogaan> nice
[19:28:08] <fuzzie> i'm pretty sure we traced all python 2.7 weirdnesses to broken environments
[19:29:59] <lynxlynxlynx> it isn't fully compatible with 2.6 like they claim, but i don't remember any issues for us either
[19:52:24] <lynxlynxlynx> aha, the dialog doesn't trigger
[19:53:39] <xrogaan> it's building ! http://pkgsubmit.mageia.org/?user=philippem
[19:54:07] <xrogaan> philipe is my mentor (i can't submit packages actually)
[19:57:46] <pupnik> oop failed
[19:58:04] <lynxlynxlynx> nice, but yeah
[19:58:07] <lynxlynxlynx> looking at the log
[19:58:37] <lynxlynxlynx> error: File not found: /home/iurt/rpm/BUILDROOT/gemrb-0.6.3-1.mga1.i386/usr/lib/gemrb/plugins/SDLAudio.so
[19:58:51] <lynxlynxlynx> sdl_mixer wasn't found, so the plugin wasn't built
[19:59:02] <lynxlynxlynx> you probably manually specified this needs to be installed?
[19:59:58] <xrogaan> yeah, I need a BuildRequires
[20:06:00] <lynxlynxlynx> the plugin is optional
[20:06:08] <lynxlynxlynx> openal is better in most cases
[20:06:47] <xrogaan> better full than incomplete
[20:07:07] <xrogaan> I mean, the user still can select which one to use, right ?
[20:07:15] <xrogaan> if both are present...
[20:07:20] <lynxlynxlynx> yes
[20:07:58] <xrogaan> There is no point to force people to use something, even if it's better methink
[20:15:50] <-- |Cable| has left IRC (Read error: Connection reset by peer)
[20:16:49] <lynxlynxlynx> jusst a buildrequires doesn't hurt, it's only for builders :=)
[20:17:17] --> |Cable| has joined #GemRb
[20:17:28] <-- |Cable| has left IRC (Read error: Connection reset by peer)
[20:19:24] --> |Cable| has joined #GemRb
[20:21:34] <lynxlynxlynx> fuzzie: just disabling BD_INTERRUPT in NIDSpecial1 didn't help at all and calling NoInterrupt before didn't change anything either
[20:22:13] <fuzzie> NIDSpecial1?
[20:22:49] <fuzzie> oh, our dialog hack
[20:23:34] <fuzzie> it's too early there
[20:24:39] <fuzzie> after the "target->ClearActions();" in DialogHandler.cpp would work
[20:24:50] <fuzzie> although it is the wrong place to do it and that ClearActions shouldn't be there anyway
[20:25:34] <fuzzie> i am buried in topology homework
[20:26:04] <lynxlynxlynx> ok, i'll try that
[20:27:07] <lynxlynxlynx> it's odd though, since in all cases, it does apply at least part of the spell, so it doesn't break on the first wait
[20:27:18] <lynxlynxlynx> could be more than one bug again
[20:27:34] <fuzzie> well, it's possible it isn't an action bug at all?
[20:29:18] <lynxlynxlynx> sure
[20:29:59] <fuzzie> but the polymorph should be obvious, if i was looking at the right code
[20:29:59] <lynxlynxlynx> there's something wrong with the output of half of those actions though
[20:33:59] <wjp> say, don't think to long on this since you're doing more useful stuff, but have you heard any reports on mysterious 50x errors with our wiki?
[20:34:03] <wjp> s/to/too/
[20:35:23] <lynxlynxlynx> nope
[20:50:45] <lynxlynxlynx> the portrait issue is separate
[20:51:12] <lynxlynxlynx> we append S/M even when not necessary
[21:06:34] <lynxlynxlynx> ok, bigger problem too, the polymorph doesn't even trigger; back to the dialog
[21:09:02] <fuzzie> so it does break on the first wait?
[21:09:19] <lynxlynxlynx> looks like it
[21:09:20] <fuzzie> and you're just seeing the effect of aerie.baf?
[21:09:31] <lynxlynxlynx> adding the nointerrupt fixed the polymorph
[21:09:59] <lynxlynxlynx> the second wait killed it again though, StartDialogueNoSet didn't get run
[21:10:21] <lynxlynxlynx> disregard that
[21:10:43] <lynxlynxlynx> it aborted due to a null object filter
[21:10:56] <fuzzie> that's .. odd?
[21:11:33] <lynxlynxlynx> -1 /even/ :)
[21:12:03] <fuzzie> oh i see
[21:12:12] <lynxlynxlynx> hehe
[21:12:17] <fuzzie> so the problem is that it's ignoring the fact there's no object.
[21:12:24] <lynxlynxlynx> the only use of this action without a parameter in the whole bg2
[21:12:46] <fuzzie> stupid dialog code :P
[21:14:04] <fuzzie> i guess a hack in the action is the easiest way to resolve that
[21:14:12] <fuzzie> but, ugh
[21:14:14] <lynxlynxlynx> it should use the last talker?
[21:14:37] <fuzzie> well, it shouldn't work at all
[21:14:51] <lynxlynxlynx> it doesn't :)
[21:14:59] <fuzzie> so i guess in the original engine it uses leftover state
[21:15:29] <fuzzie> and i have no idea what that might be, but presumably you can pick whatever works in this circumstance and add a comment
[21:15:41] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[21:15:59] <fuzzie> i mean, assuming it *does* do something in the original
[21:16:29] <lynxlynxlynx> doesn't look like fixpack touches this
[21:17:18] <fuzzie> /home/fuzzie/src/bg2fixpack/bg2fixpack/compile/soa-dlg.d:REPLACE_ACTION_TEXT_REGEXP ~\(^aerie$\)\|\(^kalah2$\)~ ~StartDialogueNoSet()~ ~StartDialogueNoSet([PC])~
[21:18:34] <lynxlynxlynx> cool
[21:18:38] <lynxlynxlynx> undocumented though
[21:19:06] <fuzzie> gosh some of the fixes there are really dumb :)
[21:19:33] <fuzzie> the comment is about "reported compile errors"
[21:19:59] <fuzzie> but some of these couldn't possibly work as-is
[21:21:04] <lynxlynxlynx> you never know what ugliness lurks in the blob
[21:23:08] <fuzzie> but i imagine the original version works fine
[21:24:42] <fuzzie> baldurdash says "AERIE.DLG - fixed: StartDialogueNoSet() to StartDialogueNoSet([ANYONE])"
[21:25:37] <fuzzie> so, maybe not.
[21:26:39] <lynxlynxlynx> it's not something obviously broken, so it is easy to imagine it slipping through qa
[21:26:48] <lynxlynxlynx> only one mistery left
[21:33:24] <xrogaan> there, gemrb is sumbitted to mageia
[21:33:47] <lynxlynxlynx> thanks
[21:34:10] <xrogaan> yw
[21:35:22] <lynxlynxlynx> i'll take the last mistery as an original bug too
[21:35:37] <lynxlynxlynx> there is no message that you lose the sword key
[21:36:01] <lynxlynxlynx> we do it in both involved actions, but they play one against the other
[21:37:14] <lynxlynxlynx> let's see if i can run it in wine
[21:49:18] <lynxlynxlynx> no luck
[21:50:36] <lynxlynxlynx> fuzzie: for that interrupt thing, do i need to reset the flag at the end or will it be taken care of soon enough?
[21:51:05] --- pupnik is now known as pupni-af-k
[21:52:35] <fuzzie> the correct way to do it is to append the relevant action at the beginning and end of what you add
[21:52:55] <fuzzie> or so i am told by Taimon, and i have no reason to doubt :)
[21:54:12] <fuzzie> i can't think of any likely relevant reason not to just do it in C++, since i think the only behaviour difference would be if ActionOverride is called before the queue is finished executing
[21:54:56] <lynxlynxlynx> ok, i'll just add a more verbose comment then
[21:55:26] --> Bo_Thomsen has joined #GemRb
[21:56:43] <fuzzie> and that code needs fixing anyway.
[21:57:17] <fuzzie> so go ahead, if it works.
[22:08:56] <CIA-36> GemRB: 03lynxlupodian * rdc44c4704d10 10gemrb/gemrb/core/DialogHandler.cpp:
[22:08:56] <CIA-36> GemRB: DialogHandler::DialogChoose: don't interrupt dialog actions
[22:08:57] <CIA-36> GemRB: fixes aerie's transformation block
[22:08:59] <CIA-36> GemRB: 03lynxlupodian * reff7703c1040 10gemrb/gemrb/core/Scriptable/Actor.cpp:
[22:08:59] <CIA-36> GemRB: Actor::SetPortrait: don't append the size suffix if it is already there
[22:08:59] <CIA-36> GemRB: such is the case with aerie and hopefully nothing has a real root ending in them
[22:10:13] <fuzzie> that is optimistic :P
[22:11:19] <lynxlynxlynx> that's me :)
[22:40:13] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[22:59:00] <-- Maighstir has left IRC (Ping timeout: 264 seconds)
[23:04:59] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:13:51] --> Maighstir has joined #GemRb