#gemrb@irc.freenode.net logs for 19 Mar 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:43:05] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[03:32:11] --> thomcom has joined #gemrb
[03:41:53] <-- thomcom has left IRC (Quit: Page closed)
[03:45:56] <-- ermo^ has left IRC (Ping timeout: 248 seconds)
[03:52:43] --> ermo has joined #gemrb
[06:50:15] --> skyphen has joined #gemrb
[06:50:56] <skyphen> Hi
[06:56:45] <skyphen> Hey I just built gemrb on OSX 10.8.3 with i686-apple-darwin11-llvm-gcc-4.2
[06:57:02] <skyphen> err g++
[06:57:32] <skyphen> anyway, im hoping to help out and contribute here. what a great project
[06:59:10] <skyphen> also installed on Galaxy Note 10.1 tablet
[06:59:31] <skyphen> Runs bg2 tob great on both platforms
[07:55:59] --> Fuzzlix has joined #gemrb
[08:14:45] --> edheldil has joined #gemrb
[08:14:45] --- ChanServ gives channel operator status to edheldil
[08:35:33] <fuzzie> good morning
[08:47:37] --> lynxlynxlynx has joined #gemrb
[08:47:37] <-- lynxlynxlynx has left IRC (Changing host)
[08:47:37] --> lynxlynxlynx has joined #gemrb
[08:47:37] --- ChanServ gives channel operator status to lynxlynxlynx
[08:49:08] <lynxlynxlynx> skyphen: great, welcome
[08:49:46] <lynxlynxlynx> if you need any guidance just ask here
[08:50:06] <edheldil> hi
[08:50:10] <lynxlynxlynx> most of us are in other timezones
[08:50:12] <lynxlynxlynx> gmornin
[09:13:14] <skyphen> thanks!
[09:13:37] <skyphen> im reading up on the src and docs so I wont bother you until later : P
[09:28:41] <-- Fuzzlix has left IRC (Quit: Bye)
[09:33:15] <lynxlynxlynx> there are many face(t)s of this engine
[09:33:32] --> Fuzzlix has joined #gemrb
[09:33:38] <lynxlynxlynx> i have to go, but if you tell us more, we can narrow your search down
[09:38:46] <Fuzzlix> Hi all :) I try to compile the sources. MinGW ok. sources ok. CMake has built the makefiles without errors. using the cmd-files from the wiki i get a compilation error :
[09:38:48] <Fuzzlix> D:\GemRB_SRC\GemRB\gemrb\plugins\Core\ActorBlock.cpp:1554:15: error: variable 'oval' set but not used [-Werror=unused-but-set-variable]
[09:38:49] <Fuzzlix> How to fix it? Thanks :)
[09:40:19] <fuzzie> woah
[09:40:22] <fuzzie> where did you get those sources?
[09:41:03] <fuzzie> they must be more than 3 years old
[09:41:39] <Fuzzlix> sources downloaded yesterday from github as a zip-file. (not using git to pull)
[09:42:06] <Fuzzlix> ok i try git :D
[09:55:12] <edheldil> fuzzie: how did you found out so quickly? :)
[09:55:32] <fuzzie> it says 'plugins\Core\ActorBlock.cpp' :P
[09:57:32] <Fuzzlix> Huh.. like ever ... Who is able to read the manuals, is in advantage ;)
[10:00:11] <edheldil> Fuzzlix: we don't have manual, so there's no unfair advantage in GemRB :-D
[10:00:25] <Fuzzlix> hehe
[10:05:24] --> fizzle has joined #gemrb
[10:19:47] <-- skyphen has left IRC (Ping timeout: 245 seconds)
[10:43:36] --> mongrol_ has joined #gemrb
[10:43:39] <mongrol_> hi folks
[10:43:50] <mongrol_> does PST widescreen work on linux?
[11:07:08] <-- WingedHussar1 has left IRC (Quit: WingedHussar1)
[11:12:53] --> rocket_hamster has joined #gemrb
[11:17:56] <Fuzzlix> no SDL_MIXER (i downloades your zip with all nessesary libs. sld_mixer is missing?)
[11:19:57] <edheldil> only for the SDLAudio plugin
[11:30:27] --> WingedHussar has joined #gemrb
[11:36:08] <Fuzzlix> D:\GemRB_SRC\gemrb_build>cmake ..\gemrb -G "MinGW Makefiles" -DCMAKE_INSTALL_PREFIX=D:\GemRB -DDISABLE_WERROR=Yes
[11:36:10] <Fuzzlix> Detected version:
[11:36:12] <Fuzzlix> CMake Error at C:/Programme/CMake 2.8/share/cmake-2.8/Modules/FindPackageHandleStandardArgs.cmake:97 (message):
[11:36:14] <Fuzzlix> Could NOT find PythonLibs (missing: PYTHON_LIBRARIES PYTHON_INCLUDE_DIRS)
[11:36:15] <Fuzzlix> (Required is at least version "2.3")
[11:36:17] <Fuzzlix> Call Stack (most recent call first):
[11:36:18] <Fuzzlix> C:/Programme/CMake 2.8/share/cmake-2.8/Modules/FindPackageHandleStandardArgs.cmake:291 (_FPHSA_FAILURE_MESSAGE)
[11:36:20] <Fuzzlix> C:/Programme/CMake 2.8/share/cmake-2.8/Modules/FindPythonLibs.cmake:186 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)
[11:36:22] <Fuzzlix> CMakeLists.txt:159 (FIND_PACKAGE)
[11:36:24] <Fuzzlix> PYTHON_LIBRARIES PYTHON_INCLUDE_DIRS versus CMake's
[11:36:26] <Fuzzlix> PYTHON_LIBRARIE PYTHON_INCLUDE_DIR ???
[11:38:37] <Fuzzlix> PYTHON_LIBRARY of course ^^
[11:58:20] <rocket_hamster> missing python?
[12:06:04] <Fuzzlix> python3.3 installed yesterday. set PYTHON_INCLUDE_DIR to my Python3.3/include and PYTHON_LIBRARY to my Python33/libs/libpython33.a
[12:06:06] <Fuzzlix> amyway no PYTHON_DEBUG_LIBRARY
[12:06:30] <fuzzie> You need python 2.x.
[12:06:49] <Fuzzlix> :) .......
[12:09:03] <Fuzzlix> yes someone wrote it in RED in the wiki not to use 3.0 -.-
[12:11:47] <lynxlynxlynx> welcome, we didn't have enough naming confusion yet
[12:12:33] <Fuzzlix> hehe ... more to come ^^
[12:19:23] <Fuzzlix> ... compiling ....
[12:42:38] <fizzle> lynxlynxlynx: have you had a chance to look at those callers of RemoveAllEffects yet?
[12:42:52] <fizzle> still looking to clean up the polymorph stuff
[12:43:15] <lynxlynxlynx> i grepped them up yesterday, don't know why i didn't look further
[13:13:56] <Fuzzlix> looks like FindLIBVLC.cmake does not like VLC V2.0.5 - version check fails.
[13:14:36] <Fuzzlix> .. or i did not put the right answers into cmake
[13:42:45] <Fuzzlix> cu ^^
[13:42:52] <-- Fuzzlix has left IRC (Quit: ChatZilla 0.9.90 [Firefox 19.0/20130215130331])
[13:58:30] <-- mongrol_ has left IRC (Ping timeout: 252 seconds)
[15:28:34] --> brada has joined #gemrb
[15:29:13] <brada> tomprince: could you remove the minimal test from mac buildbot?
[15:29:27] <brada> mac build doesnt use a cfg anylonger
[15:45:07] <fuzzie> what does it use?
[15:45:20] <fuzzie> can't we just port the minimal test to that?
[15:46:51] <brada> i can come up with something
[15:47:13] <brada> using the mac open command
[15:47:29] <brada> and giving it the path to the test data
[15:47:42] <fuzzie> i hadn't realised you were going to break this on mac
[15:47:49] <brada> ha
[15:47:58] <brada> sooo much better now
[15:48:09] <fuzzie> do you have instructions somewhere for configuring stuff for mac users?
[15:48:44] <brada> it shoudl be pretty straight forward but ill add it to the wiki and wherever else you tell me to
[15:48:54] <fuzzie> not going to tell you anything :P
[15:49:24] <fuzzie> just now we're not going to have any clue at all when mac users come here and we don't know how to explain to them how to configure CD paths and stuff
[15:49:34] <brada> its just a window with labeled options and an open dialog to locate the game
[15:50:08] <brada> im still mulling over the cd path thing
[15:50:39] <brada> i would like to make that more automatic for everybody
[15:50:43] <fuzzie> also encoding I guess?
[15:50:45] <brada> it is a huge issue
[15:50:55] <fuzzie> well, it should be automatic if you did a full install, right?
[15:51:03] <brada> not for gog
[15:51:08] <brada> its annoying
[15:51:11] <fuzzie> oh, well
[15:51:15] <fuzzie> someone with the gog releases should fix that
[15:51:17] <fuzzie> shouldn't it be trivial?
[15:51:22] <brada> fairly
[15:51:33] <brada> we could jsut add another search path probably
[15:51:34] <fuzzie> sorry, i'd assumed someone with the gog releases had done that a long time ago :P
[15:51:37] <brada> if we didnt already
[15:51:47] <brada> tbh someone may have fixed that without me noticing
[15:52:08] <brada> but if not it should be done
[15:52:27] <fuzzie> yes
[15:53:49] <fuzzie> but if people just can't configure obscure stuff then i wouldn't worry
[15:53:58] <fuzzie> just would be cool if it could be documented somewhere at some point
[15:54:03] <brada> yes
[15:54:16] <brada> techniclly they could configure any option by editing the plist file
[15:54:50] <brada> i need to add an encoding selector
[15:55:03] <brada> but i wanted to make it automatic
[15:55:10] <brada> not hard code a list
[15:55:33] <brada> but our encoding files are jsut lumped in the same dir as everything else
[15:56:16] <brada> i suppose nobody would complain if i moved them to a sub directory?
[15:56:26] <lynxlynxlynx> i changed the cachedir name in the sample configs, so for new users, there should be no gog pains
[15:56:26] <brada> but no
[15:56:32] <brada> that would complicate things
[15:56:47] <lynxlynxlynx> i'm not sure about pst though, i think you had to move some bifs there too
[15:57:04] <brada> lynx: what about the blue screen gog problems?
[15:57:11] <brada> due to needing a faux cd path
[15:57:51] <lynxlynxlynx> maybe that's the same
[15:58:09] <lynxlynxlynx> our cd paths are just extra search paths, nothing cd about them
[15:58:17] <brada> yeah
[15:58:46] <brada> at one point i had something that could get the cd paths out of baldur.ini
[15:58:53] <-- Cable_ has left IRC (Ping timeout: 256 seconds)
[15:59:09] <lynxlynxlynx> hey, we should set them to gamepath by default - that way we won't produce those confusing warnings
[15:59:19] <brada> but there is no guarantee of the being valid
[15:59:38] <brada> and usually a windows path
[15:59:45] <brada> unless we have something to convert
[16:00:01] <brada> we could intersect them with the base dir of gamepath
[16:00:17] <brada> wich i think should work great
[16:00:48] <lynxlynxlynx> what was wrong with gog in the first place, another layer deeper?
[16:01:10] <brada> iirc yea
[16:04:42] <lynxlynxlynx> with consistent naming, i bet
[16:04:55] <lynxlynxlynx> probably easier to just scan the dirs
[16:04:59] <lynxlynxlynx> bbl
[16:04:59] <-- WingedHussar has left IRC (Quit: WingedHussar)
[16:08:38] <brada> yeah but if you wanted to get rid of the path warnings...
[16:09:49] <brada> grrr when i renamed that file i failed to remember we already had a default.ini in the same place
[16:15:47] <rocket_hamster> took an estimated 40 years of effort (COCOMO model)
[16:15:56] <rocket_hamster> says ohloh
[16:15:57] <rocket_hamster> o.O
[16:17:30] <fuzzie> COCOMO model assumes it's finished, mind :p
[16:23:23] <psch> brada: i didn't have any issues with config file saving
[16:23:34] <psch> i tested it when it came up the last time here in chat, about a week ago i think?
[16:23:46] <psch> no idea if you changed anything, but i don't really have time to look right now
[16:23:58] --> |Cable| has joined #gemrb
[16:49:56] <lynxlynxlynx> and it's a wierd model, it's been saying 40 years for the past few years
[16:57:30] <rocket_hamster> yeah maybe its kind of logarithmic
[16:57:50] <rocket_hamster> and you cant esceed certain threshhold
[16:58:05] <rocket_hamster> that would be good for paying for long projects and saving money for management
[16:58:13] <rocket_hamster> :D
[17:19:12] <-- brada has left IRC (Quit: brada)
[17:42:06] <-- fizzle has left IRC (Ping timeout: 264 seconds)
[17:46:41] --> brada has joined #gemrb
[18:02:17] <-- Drakkar has left IRC (Read error: Connection reset by peer)
[18:02:48] --> Drakkar has joined #gemrb
[18:19:55] --> Fuzzlix has joined #gemrb
[18:23:02] <-- rocket_hamster has left IRC (Quit: bye!)
[18:28:05] --> Yoshimo has joined #gemrb
[18:57:49] <-- brada has left IRC (Quit: brada)
[19:25:10] --> brada has joined #gemrb
[19:51:51] <lynxlynxlynx> Bad Line in file: defaults.ini, Section: [Program Options], Entry: ' ; Program Options - Audio'
[19:52:12] <lynxlynxlynx> looks like we don't like spaces before comments?
[19:54:32] <lynxlynxlynx> config saving still works fine on linux (with the rewrite)
[20:37:42] <-- Yoshimo has left IRC (Quit: Yoshimo)
[20:43:11] <-- nutron has left IRC (Quit: I must go eat my cheese!)
[20:49:02] --> rocket_hamster has joined #gemrb
[20:53:30] <lynxlynxlynx> woah, wtf happened
[20:53:45] <lynxlynxlynx> can't run bg2 or iwd2 anymore, paths get omnomnomed
[20:53:54] <lynxlynxlynx> how and the demo work
[20:54:34] <lynxlynxlynx> something doesn't like spaces it seems
[20:57:14] <fuzzie> it's all totally broken for me
[20:57:17] <lynxlynxlynx> commenting out cdx entries makes it load, but data is still missing
[20:57:35] <fuzzie> i thought brada was going to fix my gemrb.cfg path, but it's broken still
[20:57:35] <lynxlynxlynx> at least i don't get a ctd anymore
[20:59:30] <fuzzie> with a completely useless error msg, too :/
[21:01:02] <fuzzie> i seem to have gained a broken clock overlaying my sleep button, too
[21:01:05] <fuzzie> in bg2
[21:01:49] <lynxlynxlynx> fizzle was playing with that
[21:01:57] <fuzzie> and i have identify in my spells list
[21:02:03] <fuzzie> i mean, the actionbar one
[21:02:12] <lynxlynxlynx> maybe he removed the controlid check
[21:02:17] <lynxlynxlynx> identify is normall
[21:02:23] <lynxlynxlynx> unless you can click it
[21:03:44] <fuzzie> hm, right, avenger already o.Oed about it, I see in log
[21:04:04] <fuzzie> i don't think it was like that in original but my memory is pretty bad, so i'll take your word for it :)
[21:04:15] <lynxlynxlynx> hmm, interesting pattern, the last 14 chars of the path remain (out of 35)
[21:06:00] <fuzzie> i don't understand fizzle's "better feedback when learning spells fails" commit, since that commit seems to remove the check for whether it failed...
[21:07:39] <fuzzie> where do the paths come from?
[21:07:50] <fuzzie> brada's new INIConfig::InitWithINIData does weird things with spaces
[21:07:56] <lynxlynxlynx> it's the datapath
[21:08:22] <lynxlynxlynx> but it is broken for a nonspacey path too
[21:09:02] <brada> o_O
[21:09:16] <brada> iirc i just moved code from one place to another there
[21:10:03] <lynxlynxlynx> http://paste.debian.net/242924/
[21:10:06] <Seniorita> debian Pastezone
[21:10:40] <lynxlynxlynx> cdx that are autogenerated are fine, the others get their path chopped
[21:10:44] <fuzzie> ah, yes
[21:10:57] <fuzzie> brada: I can't follow that very easily if you add in one commit and remove in another, sorry :/
[21:11:19] <fuzzie> especially since you changed the code
[21:11:50] <lynxlynxlynx> hah, sf added a git hook to advertise the new forge
[21:12:00] <fuzzie> seriously?
[21:12:04] <wjp> yes
[21:12:18] <wjp> I've heard other people complain on #sourceforge too...
[21:12:27] <fuzzie> if they want people to migrate then how about they clear their ticket queue first?
[21:12:44] <fuzzie> sigh
[21:14:05] <fuzzie> what on earth is going on in the CD loop in the config code?
[21:14:19] <fuzzie> my head explodeth
[21:15:23] <wjp> hm, since I'm just staring at a progress bar anyway, let's see
[21:18:16] <fuzzie> I thought I'd just look at the diff but I can't quite follow it atm.
[21:28:55] <wjp> now I'm not sure what ResolveFilePath does, but why is that mixed inside a strtok run?
[21:31:24] <fuzzie> i assume the idea is to split the path up into components and then add all components
[21:31:59] <fuzzie> which seems like, well, a really bad idea in any case, but the code has been there for years
[21:32:35] <wjp> not the ResolveFilePath call inside the strtok loop, though
[21:33:07] <wjp> unless I'm misreading this diff of course, which is entirely possible
[21:33:33] <fuzzie> yes, I just mean the strtok run
[21:34:37] <fuzzie> the ResolveFilePath is meant to be passed a mutable buffer of course
[21:34:55] <fuzzie> so you'd think the const would deal with that
[21:35:03] <fuzzie> but I guess the diff is also casting the const away...
[21:35:27] <fuzzie> so I guess that could easily be corrupting the heap right there
[21:36:00] <fuzzie> we've also gone from stricmp to a case-sensitive hashmap which might break some configs
[21:39:24] <fuzzie> i guess that should surely not be the culprit unless lynx has '~' in the path
[21:39:44] <lynxlynxlynx> i do
[21:40:02] <fuzzie> well, then brada broke it.
[21:40:11] <fuzzie> no mystery :)
[21:40:20] <brada> yeah i thought we already knew that...
[21:40:38] <fuzzie> this is kind of impossible to review because you changed way too much code when moving it
[21:41:21] <wjp> yes... even with three windows (old, new, diff) it's not very clear what changed
[21:42:21] <fuzzie> lynxlynxlynx: so which path exactly breaks?
[21:43:08] <fuzzie> if it's a CD path then it's because, as wjp noted, ResolveFilePath is now operating on the internal buffer of std::string on Interface.cpp:1587, rather than the CD[i] one as it was before
[21:43:09] <lynxlynxlynx> cd1 and cd2 http://paste.debian.net/242924/
[21:43:12] <Seniorita> debian Pastezone
[21:43:40] <wjp> (not to mention that strtok is breaking that same internal buffer)
[21:43:54] <lynxlynxlynx> both of them use ~
[21:44:01] <fuzzie> presumably moving the ResolveFilePath line one line down, and making it operate on CD[i][CD[i].size()-1] will deal with that
[21:44:12] <fuzzie> since I assume that will then use the nice std::string variant
[21:44:25] <fuzzie> the ModPath loop presumably has the same bug
[21:44:33] <wjp> or just manually put it in a string
[21:44:44] <wjp> but I'm not happy with running strtok on that either
[21:45:07] <fuzzie> but yes, running strtok on a const char * is evil inherently
[21:45:17] <fuzzie> i'm just thinking of a temp fix to get lynx running again
[21:45:25] <wjp> right
[21:46:22] <lynxlynxlynx> no need to hurry
[21:46:40] <lynxlynxlynx> i wanted to check what was new, mission accomplished >>
[21:48:07] <fuzzie> well, from my perspective, it no longer errors when it can't find a config file, which has broken my brain
[21:48:14] <fuzzie> but also no hurry
[21:57:20] <-- rocket_hamster has left IRC (Quit: bye!)
[22:01:57] <brada> fuzzie: should look for ./GemRB.cfg again
[22:05:57] <fuzzie> yes! shiny, thanks
[22:06:13] <brada> yeah sorry for breaking it
[22:06:17] <brada> thought it was a hack
[22:06:19] <brada> :p
[22:06:22] <brada> and the other stuff
[22:06:41] <brada> need to run to school
[22:07:39] <-- brada has left IRC (Quit: brada)
[22:07:54] <fuzzie> is there anything in gemrb which isn't a hack, really? :)
[22:09:35] <fuzzie> tricky spellhold bits seem to mostly still function
[22:11:32] <fuzzie> [Python/ERROR]: Traceback (most recent call last):
[22:11:32] <fuzzie> [Python/ERROR]: File "../gemrb/GUIScripts/InventoryCommon.py", line 756, in OpenItemWindow
[22:11:33] --> mongrol_ has joined #gemrb
[22:11:35] <fuzzie> [Python/ERROR]: ResRef = slot_item['ItemResRef']
[22:11:38] <fuzzie> [Python/ERROR]: TypeError: 'NoneType' object is not subscriptable
[22:11:51] <fuzzie> ^- second opening of a container after taking stuff in/out of one
[22:16:25] <-- mongrol_ has left IRC (Client Quit)
[23:36:22] --- Fuzzlix is now known as Fuzzlix__
[23:37:20] <-- lynxlynxlynx has left IRC (Remote host closed the connection)