#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2013-01-13

Timestamps are in GMT/BST.

00:00:22kaenQuartz, yeah that's right
00:00:26kaenjust tested it
00:00:28martin_bedeit took a while to compile it
00:00:30QuartzBut where in the level line?
00:00:35kaenat the end
00:00:45kaenTurret 1 -0.2 -2 0 W=Seeker
00:00:54QuartzWhy is that not working for me. Gahhh
00:00:58QuartzAlright I'll give it another shot anyway
00:01:07kaenQuartz probably not in the 018 release
00:01:18kaenI think it has to be on a patched server like the contest server
00:01:19QuartzJust checked I am
00:01:24sam686"W=Heat Seeker" ? (with quotes)
00:01:26Watusimotoso this line crashes, right? ::: mOwner->setVert(dest, 1)
00:01:32martin_bedeyes
00:01:33Watusimotoand mOwner is not NULL
00:01:38martin_bedeyes
00:01:40Watusimotodest can;t be null
00:01:43kaensam686, no it's just seeker. and no quotes.
00:01:45Watusimoto1 isn;t null
00:01:46Quartz..okay randomly it just worked
00:01:52kaenoh good :)
00:01:56Watusimotoso what's left to crash?
00:02:08QuartzIs it Burst or Burster, I forget
00:02:13kaenBurst
00:02:16QuartzCool
00:02:35kaenQuartz, http://bitfighter.org/luadocs/group___weapon_enum.html
00:02:38martin_bedei don't know. this bug is like a curse. i don't even know how it was introduced
00:02:38Watusimotook, try this
00:02:43kaenthere's the list
00:02:59Watusimotowell, it is odd that it affects only you
00:03:07WatusimotoI don;t get it, raptor doesn;t get it
00:03:17bobdaduck has joined
00:03:20Watusimotoand none of the relevant code has been touched for a while
00:03:30Watusimotoso I would suggest testing the following:
00:03:35Watusimoto1) do a clean rebuild
00:03:43martin_bedealready done
00:03:53Watusimoto2) checkout a fresh copy of the code in the repo, and try that
00:04:02Watusimotoin a idfferent folder
00:04:36martin_bedei'm sure that would work, but all of our work would be lost then
00:04:47Watusimotonot at all
00:04:48bobdaduck Quit (Client Quit)
00:04:57Watusimotoif you do it in a different folder, nothing is lost
00:05:03Watusimotothen you can compare the two code bases
00:06:05Watusimotoand, if it comes to that, you can use diff to move your changes from the broken code base to the working one
00:06:29Watusimotoor if the fresh checkout is broken, we can try something different
00:07:02martin_bedei had to manually merge once, probably i did something wrong and it showed now
00:07:04qwebirc373743 has joined
00:07:08qwebirc373743sorry i crashed
00:07:15qwebirc373743 has left
00:08:12amgine1234567890 Quit (Ping timeout: 245 seconds)
00:08:37Quartz Quit (Ping timeout: 245 seconds)
00:10:35Watusimoto Quit (Read error: Connection reset by peer)
00:11:56Watusimoto has joined
00:13:50martin_bedeWatusimoto: fixed it :D
00:14:01Watusimotohoe
00:14:04Watusimotohow
00:14:35martin_bedei put back the init in the teleporter's constructor
00:14:51martin_bedesomehow i got lost when i tried adding the destinaton paramter
00:14:59martin_bedeit*
00:15:02Watusimotoha!
00:15:31WatusimotoI figured something like that had happened, but of course I had no idea where to look
00:15:37Watusimotohence the clean codebase
00:15:53Watusimotobut I alwo was beginning to think your cpu was broken :-)
00:16:53martin_bede:)
00:17:56martin_bedeso what should i pass to a wallitem in lua?
00:18:21martin_bedei get a couldn't validate params currently
00:18:37martin_bedeusing the following
00:18:37martin_bedelocal j = WallItem.new(100,100, 111,111);
00:19:17Watusimotolet me see... I would think that would work
00:19:38martin_bedebut it wants a geom and an int
00:22:15martin_bedeif i add another number i get a segfault
00:22:18martin_bedeis that progress?
00:22:49martin_bedeprobably there is something wrong with the constructor
00:23:57Watusimotooh right, it needs a width
00:24:05Watusimotoso yes, that's progress
00:25:06Watusimotomartin_bede: I am falling asleep... so I'm going to go in about 3 mins... if you need anything else tonight, speak now
00:25:34martin_bedeokay, i would like some suggestions for fixing that then
00:25:38martin_bedeWallItem
00:25:47Watusimotolet me see your constructor
00:26:13martin_bedehttp://pastebin.com/HBaxvKYJ
00:27:27martin_bedei think setPos(getPointOrXY(L, 1)); should be replaced with something like setNewGeometry(geomPolyLine);
00:27:33martin_bedebut i dopn't know waht exactly
00:27:46YoshiSmb has joined
00:27:50martin_bededon't; what*
00:27:58Watusimotomy guess is the setPos is the problem
00:29:12WatusimotogeomPolyLine is just an enum; that method won't help
00:29:18YoshiSmb Quit (Client Quit)
00:29:35Watusimotoso when I get stuck, I look at processArguments
00:29:47Watusimotobecause that is doing something simlar, with args from a level file
00:30:10Watusimotothat calls readGeom()
00:30:57Watusimotowhich calls PolylineGeometry::readGeom
00:31:11Watusimotowhich calls static void readPolyBounds
00:31:22Watusimoto(does anything do any work around here, or just call something else?!?)
00:31:42Watusimotoso I think you need this:
00:31:44Watusimotovoid PolylineGeometry::setGeom(const Vector<Point> &points)
00:32:10WatusimotoI think you need to get the points from Lua into a vector
00:32:15Watusimotothen pass them to this method somehow
00:32:27martin_bedethanks
00:32:54Watusimotothen call onPointsChanged();
00:33:17Watusimotooh look
00:33:18WatusimotoS32 WallItem::setGeom(lua_State *L)
00:33:55martin_bedeit does this?
00:34:14Watusimotosomehting similar
00:34:24Watusimototracing that through to where work is done, I find this:
00:34:26Watusimoto Vector<Point> points = getPointsOrXYs(L, stackIndex);
00:34:26Watusimoto GeomObject::setGeom(points);
00:34:26Watusimoto onPointsChanged();
00:34:38Watusimotowhich looks like exactly what you need
00:35:03Watusimotoso I think you should monkey with that a bit
00:35:10martin_bedebut first i have to remove the width from L
00:35:31Watusimotoyes, you probably do
00:35:51Watusimotoit will be the last(?) item on the stack
00:36:08martin_bedeit is
00:36:10Watusimotowhich has an index of -1
00:36:30Watusimotoso you;ll do lua_getInt(L,-1) or something
00:37:00Watusimotothen you'll need to remove it from the stack using (maybe) lua_pop
00:37:15Watusimotonot sure if that's the right method, but something similar; easy to google
00:37:27Watusimotothen you should be set
00:45:02raptor checks the logs
00:45:08raptor runs away screaming
00:45:17raptorok, i do not want to read... anything i miss?
00:51:55martin_bedei didn't notice anything very important
00:58:58martin_bederaptor: Watusimoto told me to look at processArguments when I'm stuck, but it's not implemented yet even there, what am i supposed to do?
00:59:42martin_bedei'm talking about the Zone class
01:00:00raptorprocessArguments might be part of the parent class
01:00:27raptorfollow the parents up until you find a class that has it
01:00:43martin_bedethere is a processArguments there, but it leads to unimplemented methods
01:00:48martin_bedethat's what i meant
01:00:54martin_bedeso it uses unimplemented stuff
01:01:33martin_bedereadGeom() just throws an error
01:02:36martin_bedeTNLAssert(false, "Not implemented");
01:04:58raptorok, let me look...
01:06:27raptorZone.cpp:85
01:06:30raptori see Zone::processArguments(S32 argc2, const char **argv2, Game *game)
01:07:52martin_bedei noticed it too, but it calls readGeom
01:09:55martin_bedenevermind, it's implemented for poligons
01:11:18raptorbasic geometry is determined by what type of geom object it is
01:11:24raptorlike point or line or polygon
01:11:43raptorWatusimoto: you are asleep now, yes?
01:13:34YoshiSmb has joined
01:13:46martin_bedehe went to sleep a few minutes ago
01:15:15YoshiSmb Quit (Client Quit)
01:27:36Watusimotoyes
01:27:39WatusimotoI mean no
01:27:43Watusimotoor rather yes
01:27:59Watusimotoso...
01:28:03Watusimotogoodnight!
01:28:13martin_bedejust a sec
01:28:13Fordcars has joined
01:28:21Watusimotoyes?
01:29:00martin_bedethis zone thing, any easy way?
01:29:36Watusimotoyes
01:29:41Watusimotojust like loadoutZone
01:29:48Watusimotozone is the parent of loadoutZone
01:30:00martin_bedeok thanks, bye
01:30:15Watusimotonight
01:34:08raptornight
01:35:00Watusimoto Quit (Ping timeout: 255 seconds)
01:40:00-moorcock.freenode.net- *** Looking up your hostname...
01:40:00-moorcock.freenode.net- *** Checking Ident
01:40:11-moorcock.freenode.net- *** No Ident response
01:40:28-moorcock.freenode.net- *** Couldn't look up your hostname
01:40:36BFLogBot has joined
01:40:36Topic is 'Bitfighter 018 released! | http://www.bitfighter.org/downloads | Welcome GCI students! Please introduce yourself and feel free to ask any questions.'
01:40:37Set by raptor!~raptor@unaffiliated/greenmachine on Thu Dec 06 10:11:09 GMT 2012
01:40:37-ChanServ- [#bitfighter] Welcome to #bitfighter. This is an IRC channel, many or all of the users may not be paying attention. Please have patience when waiting for a response.
01:40:51martin_bedemy clone contains every change, a link is on the task page
01:41:44martin_bedei think we tested/fixed everything with Watusimoto
01:42:29martin_bedethere are 2 missing parts but those are intentional, written about in the code (comments) and were discussed with him
01:44:05martin_bedebye
01:44:15martin_bede Quit (Quit: Bye)
01:47:17raptorhi Fordcars myfault!
01:47:22raptorthey should be gone now...
01:47:38Fordcarsoh ok sorry!
01:51:18Fordcarshey did you get a notification app to work on mac yet?
01:52:34raptor Quit ()
02:13:59Fordcarsok....
02:34:33raptor has joined
02:34:33ChanServ sets mode +o raptor
02:35:07raptorsorry Fordcars, internet problems... no, no notifier for mac yet...
02:35:12raptorbe back later..
02:47:54Nothing_Much Quit (Ping timeout: 255 seconds)
03:04:48Nothing_Much has joined
03:13:20Nothing_MuchHow's the Desura application for this game going so far?
03:19:22kaenHaven't asked watu and I think he's sleeping now
03:21:50raptorhi
03:22:09raptori think he submitted it, but it requires some stuff - like screenshots, video, etc..
03:22:24raptorso it isn't approved or anything, by a long shot
03:27:24raptornow to see if the latest commits compile...
03:28:55raptorhey kaen, I'm thinking your cmake system is probably ready to be added to bitfighter - was there anything left to do on it?
03:32:04kaenI have to round up the requisite find modules and use them.
03:32:22kaenI've already fixed the logging issue, but after a good once over they'll be ready.
03:33:44kaenis there some sort of global for a client's local ClientGame?
03:34:07Nothing_MuchYeah I know, he said that he already submitted an application, I was wondering if there was a status update
03:34:54kaenoh, gClientGames looks promising
03:37:56Fordcarsok thanks
03:38:11raptorkaen: gClientGame...
03:38:12raptorbut
03:38:27raptori know watusimoto has sworn in his wrath to get rid of the global Game objects
03:38:41raptorbut he also keeps using them in things...
03:38:58kaenI'm glad he's sworn such a thing
03:39:29FordcarsI think I might have managed to download the .json on bitfighter.org with XMLhttpRequest
03:39:32raptoryou'll find three ways to grab the Game object: gWhateverGame global, mGame object member, getGame() method
03:39:46raptorall used in various ways all over the place..
03:39:57kaenawesome \o/
03:41:14BFLogBot Commit: 4d186156b2cf | Author: buckyballreaction | Message: Ability to run anonymous master connections with a callback is more or less complete. This paves the way for being able to start bitfighter without requiring a master connection. We can then move the UINameEntry screen to another location..
03:42:20raptor^^ that was my last week's work...
03:42:39raptori know i know.. it's not much
03:45:22raptorbe back later
03:49:43kaenFordcars, you do that every time you go on the website ^^
03:50:04Fordcarslolololololololololol nah dude!
03:50:11kaen...
03:50:19kaenno really
03:50:22kaenit uses XHRs
03:50:37kaenthat's how it auto-updates
03:50:51FordcarsI was working on a notifcatipons app thing that would say how many players are on Bitfighter etc
03:51:32kaengood news!
03:51:42kaenthere are already three of them :)
03:51:42FordcarsTo go on bitfighter.org, I use something called.... wait for it.... a web browser 0.o
03:51:59FordcarsYah I was working for aMac dashboard app
03:52:04kaenoh good
03:52:06Fordcars*yeah
03:52:11kaenthat's the only one there isn't :)
03:52:15Fordcarslol yeah
03:52:43kaenthere's a python one that might work but I don't know a thing about mac or how to fill its deps
03:53:06kaenI'm working on using GL shaders in bitfighter...
03:53:09kaendon't tell wat
03:53:12Fordcarslol ok
03:53:36Fordcarsgl shaders? was that only for 3d graphics?
03:53:50kaennope
03:53:54Fordcarsah ok
03:54:02kaenyou can do all sorts of things with shaders
03:54:09Fordcarsyeah
03:54:11kaenimo the 2d effects are cooler since they're less costly
03:54:23Fordcarsyeah
04:24:50raptorkaen: speakest thou of shaders??
04:25:07raptordoes this mean you're working on porting to GLES 2?
04:38:59kaenuh...
04:39:00kaennope.
04:39:14kaensorry :<
04:39:16raptorefficiency project?
04:39:19kaennope
04:39:22kaeneye candy
04:39:23kaenI'm lame.
04:39:26raptoror... yes that..
04:39:31raptor:)
04:39:47kaenI have high hopes for it
04:39:53kaenthe shader framework is compact and super easy to use
04:40:02kaenI put in some xmacros...
04:40:06raptornoooooo
04:40:25kaenthey're a workable solution to a legitimate problem
04:40:37kaenthe downside is small compared to the gains
04:41:02raptorare you using GLSL?
04:41:05kaenyeah
04:41:06kaenbut
04:41:12raptorso we'd need openGL 2.0..
04:41:20kaenI'm using extension testing and runtime function lookup via sdl
04:41:24kaenno we don't!
04:41:33kaenbecause it can intelligently disable itself at runtime
04:41:38kaenor even be ifdef'd out
04:41:38raptorahh
04:41:41raptorcool
04:41:44kaenI've done it with 0 external libs
04:41:52kaendidn't even have to touch the linker flags
04:42:30raptorWOW
04:42:31raptorcool
04:42:35kaenit's also got some nice hooks for an options interface
04:42:36raptorso now the question is...
04:42:42raptorwhat eye candy?
04:42:46raptoror maybe i should be patient..
04:42:50kaenanything you can dream of
04:42:55kaenbut yeah be patient
04:42:58kaentoday was productive
04:43:10raptorheh, ok
04:43:20kaenit's literally day one of me working with glsl, matrix operations, 3d processing, and a few other things
04:43:43raptorto be honest, i have little knowledge of how shaders actually work... i've been reading up on it since you've mentioned it
04:44:05kaenthey're pretty straight forward
04:44:06raptori just knew that they run directly on the GPU
04:44:17kaenit's hard finding information about it for a consistent version
04:44:25raptorthey're already part of the rendering pipeline, right?
04:44:31kaenoh yeah
04:44:39kaenthey replace parts of the pipeline
04:44:40raptorit's just that now you can program them directly with GLSL
04:44:51raptorif i understand right enough..
04:44:56kaenyes
04:44:58kaenspecific parts
04:45:12kaenthe vertex and fragment shaders in the version I'm targeting right now
04:45:26kaenwhich lets you move vertices and change pixel colors
04:45:40kaenbased on inputs from userspace and default rendering pipeline behavior
04:46:14raptorright now we have all the vertices with GL ES 1
04:46:30raptorso now, you directly alter them at the next step?
04:46:44kaenI can, but for now I just pass through
04:47:03raptorok so
04:47:03kaenI'm mostly interested in the fragment shader, which will i.e. let me draw a procedural foggy cloud for explosions
04:47:07kaeninstead of dots
04:47:12raptorooooo
04:47:33kaenglsl has builtin noise functions
04:47:37kaenit's so beautiful
04:47:39raptorso instead of doing the math for that on the CPU (in our program), you can let the hardware do it?
04:47:42kaenyes
04:47:50raptorgreat! i'm starting to understand..
04:48:09kaenalso consider, specialized hardware for vector and matrix operations
04:48:11raptorso theoretically, we could *increase* eye candy, yet *decrease* CPU?
04:48:14kaenyou get access to that in glsl
04:48:17kaenyes.
04:49:14raptorsweet!
04:49:36raptorthings are making so much more sense now... like all those weird options in games to turn on different features
04:49:44kaenexactly!
04:51:04kaenin fact it's all up and running and ready for expansion right now... except I can't translate game points to "window space"
04:51:18kaenand we really need that to do anything interesting...
04:51:36Fordcarsgeez kaen did you get like an OpenGL training????
04:51:43kaenno
04:51:44kaenI read
04:51:45kaenand read
04:51:54kaenand when my eyes hurt I went for a walk
04:51:57kaenand came home and read.
04:52:01sam686Such opengl fragment shaders isn't very useful without textures..
04:52:01Fordcarswow
04:52:02kaenand then programed
04:52:10kaensam686, true
04:52:21kaenbut if we only need them for shaders, it's really moot
04:52:24Fordcarsbut we won't put textures, right?
04:52:35kaenwe won't need to package textures
04:52:36raptorread and read, walk, read... that's what i did when converting to SDL!
04:52:46Fordcarslets keep it vector
04:52:50kaenwe can render to textures and manipulate those
04:53:00Fordcarswow ok
04:53:05kaenthe data never leaves the gpu, it's also very speedy
04:53:11kaenI test on a 64mb video card...
04:53:46YoshiSmb has joined
04:53:58sam686what about my Radeon 9200 128 MB that only has Opengl 1.4? it doesn't have fragment shaders or anything on Opengl 2.0 or later..
04:54:08raptorhi sam686
04:54:24raptorkaen will do sneaky stuff, sam686, to not turn that on...
04:54:27Fordcarsok so it like renders a texture (ex: smoke) and then you can manipulate it?
04:54:36kaenif it really doesn't (my earlier model does support it...) it'll automatically disable itself
04:54:41kaenFordcars, exactly
04:55:02kaenI have Radeon 4100
04:55:06Fordcarsnice whoa dude I actually understood something in less than an hour1!!!
04:55:29kaenwhat's the output of glxinfo for you, sam?
04:55:36kaenrather:
04:55:42kaenglxinfo | grep shader_
04:56:16sam686texture... like smoke? http://www.youtube.com/watch?v=U4z4lwCRSu4 (been a long time since I made that video)
04:57:06kaenheh sure if you wanted
04:57:09kaenthat's really common
04:58:29FordcarsI still only have 32mb vram on my "coding" computer
04:58:34Fordcars:/
04:59:23kaenI dream about getting a computer with an expensive video card some day
04:59:31kaenplaying games at maximum graphics settings...
04:59:58Fordcarswith 12 cores
05:00:09Fordcarsand a huge 42 inch hd screen
05:00:17Fordcarsyeah
05:00:21raptorevery time you get a computer with an expensive video card... you want another one in a month
05:00:38raptorok maybe 6 months..
05:01:17Fordcarsyep. But I have had the same one for like 7 years lol
05:01:53FordcarsThe graphics in the games these days are so intense
05:05:29YoshiSmb has left
05:11:41raptorlooks like nvidia has a whole host of non-standard shaders you can use..
05:18:09amgine1234567890 has joined
05:18:32amgine1234567890werid was anythign chaged with BF it was working for me but now when i try to run it it comes up full screen and gets frozen
05:18:43amgine1234567890at the menu screen
05:20:39amgine1234567890yeah this is the third time now =p
05:20:44raptorsam686: can you connect to master with normal 018?
05:20:44amgine1234567890ive tried running BF
05:20:59raptoramgine1234567890: i recently updated the master server.. maybe i broke something
05:21:33amgine1234567890the BF i have (which was the first release BF 18) gets frozen at the menu screen where you login and is in full screen whne i try to run it
05:22:41raptorkaen or Fordcars, can either of you start 018 and connect to master like normal?
05:23:49amgine1234567890however i have the original BF 18 release so i may be out of date so im not compatible anymore
05:24:03Fordcarsok
05:24:09raptorno... we take care to make sure everyone is still compatible
05:24:21raptoror at least.. we try really hard..
05:24:25amgine1234567890*please not work for fordcars please not work for fordcars*
05:24:36Fordcarshey :/
05:24:45Fordcarsbrbrbrbrbrbrbrbrbrb
05:24:51amgine1234567890what im hoping its not my version XD
05:25:26amgine1234567890please not work please not work please not work
05:25:33amgine1234567890sorry spamming =p
05:25:42amgine1234567890 *bans amgine for 60 seconds*
05:26:35amgine1234567890*waits*
05:26:56amgine1234567890i keeps freezing for me
05:27:10amgine1234567890 can somon else try it other then me
05:27:19raptoramgine, there are 4 of us in playing right now
05:27:21raptorincluding Fordcars
05:27:25amgine1234567890...
05:27:32amgine1234567890what happened to my version =p
05:27:55amgine1234567890can you give me the newest working zip build
05:30:06amgine1234567890*tries exctracting and replaceing all files from my zip*
05:30:47amgine1234567890let me try somthing
05:30:52amgine1234567890 Quit (Quit: Page closed)
05:42:04amgine1234567890 has joined
05:42:07amgine1234567890nope i cant get it working
05:42:31amgine1234567890oddly enough BF is set to run in windowed in hte ini but it automaticly runs full screen
05:42:47amgine1234567890correction full screen strached
05:52:17amgine1234567890note commit chage whne score is set to 1000000000 or greater to read Unlim
06:04:48Fordcars Quit (Quit: This computer has gone to sleep)
06:25:07amgine1234567890 Quit (Quit: Page closed)
06:33:27raptor has left
06:45:43Darrel has joined
06:55:56martin_bede has joined
06:56:13martin_bedehi
07:54:48mayanks43 has joined
09:42:35Watusimoto has joined
09:46:25martin_bedehi Watusimoto
09:47:00Watusimotohi
09:51:50Watusimotohow'd it go last night?
09:53:08martin_bedeit's completed, i think
09:53:15Watusimotoexcellet
10:01:10martin_bedewhen will you have time to review it? i can't claim a task after 18:00 and i would like to do all of these parameterizing tasks
10:04:38WatusimotoI'm looking at another task at the moment... I'll try to get to it in about 1/2 hour
10:04:54Watusimotothe review should be pretty quick, as I know what I'm looking for
10:08:18martin_bedegreat, thanks
10:10:16Watusimotoso everything worked except teleporter and turret, right?
10:11:52martin_bedeyes
10:12:19martin_bedethey kinda work but not too logically
10:13:16martin_bedei didn't check zone much though, couldn't figure out what to logprint
10:13:44martin_bedeanyway i did it they way you recommended, so there shouldn't be a problem
10:15:21LordDVG has joined
10:21:11Watusimotomartin_bede: where is your code?
10:22:11martin_bedein the clone
10:22:21martin_bedea link was posted to the task page
10:23:31Watusimotoodd; I didn't get the usual email
10:24:30Watusimotocan you make a diff from the main codebase? that will make it easier to review
10:25:06martin_bedehg diff <linktomain>?
10:25:23Watusimotoprobably :-)
10:32:09martin_bedenope that doesn't work. i also tried outgoing, but that just prints the commits to be pushed
10:35:22Watusimotohg diff -r <your head changeset id> -r <other dev changeset id> . ??? well, there is the brute force way... download another copy of the codebase and diff the whole thing...
10:35:35WatusimotoThat from here: http://stackoverflow.com/questions/3559154/how-can-i-see-a-list-of-all-files-that-are-different-between-two-hg-repositories
10:42:37martin_bedejust did the second
10:42:53martin_bedefor 10 mins i wondered why output to text file doesn't work
10:42:59martin_bedei was looking at the wrong dir
10:43:06LordDVG Quit (Remote host closed the connection)
10:44:04martin_bedehttps://docs.google.com/file/d/0B9bwI2wcChsIWUswVi1hb3B0REk/edit
10:44:10martin_bedecouldn't upload it to task page
10:44:15martin_bedeso i uploaded it there
10:45:29WatusimotoI don't have access to that... but... please use pastebin/pastie if you can!
10:45:39Watusimotothey're so much easier to deal with
10:45:47martin_bedeokay, but it's a quite large file
10:45:58Watusimotoit's all text, right?
10:46:01martin_bedeyep
10:46:05Watusimotothen no problem!
10:46:17Watusimotobut it shouldn't be thaaat large, right?
10:46:29martin_bedeit shouldn't be, but it is
10:46:32Watusimotoit only has your edits in it, I hope
10:46:39martin_bedesomehow other people's commits got in
10:46:41martin_bedetoo
10:46:49Watusimotowell, let's have a look
10:47:01Watusimotomaybe you need to update your codebase
10:47:16Watusimototo get others' edits out
10:47:28martin_bedehttp://pastebin.com/tiUCWKKe
10:48:14martin_bedethe code is the latest revision in both repos
10:48:32Watusimotook, first off, that's not too big for pastie/whatver, so don't worry about that
10:50:01Watusimotodo you know when you made your copy of the repo (the one you were working off)?
10:50:09Watusimotoie. do you know the parent version?
10:50:33Watusimotobecause you could try making the diff off of that
10:50:48Watusimotoyou have my work from last night in there :-)
10:51:14martin_bedei don't really know that, how can i figure that out?
10:51:29martin_bede(the parent ver)
10:53:00Watusimotonot sure. (I struggle with these tools too!)
10:53:08WatusimotoI'm looking at your google code repo
10:53:26Watusimotoit seems to have at least fairly recent changes from me in it
10:53:59Watusimotoalright, let me try something
10:57:27BFLogBot Commit: 4e2e85b73bf1 | Author: watusimoto | Message: Only reorder if we have args that require it
10:57:28BFLogBot Commit: f5055a22deed | Author: watusimoto | Message: Minor
10:57:30BFLogBot Commit: 07d24265b35a | Author: watusimoto | Message: Comment
10:57:31BFLogBot Commit: 4a83d05a4578 | Author: watusimoto | Message: Rename luavec macro to make code more readable
10:57:33BFLogBot Commit: 904d8657464e | Author: watusimoto | Message: Mitsov announce patch
10:57:34BFLogBot Commit: a22ba9555f27 | Author: watusimoto | Message: Comments
10:57:36BFLogBot Commit: 70cdfe0ad762 | Author: watusimoto | Message: Merge
10:59:48Watusimotook, this patch isn't going to work
11:00:16WatusimotoI tried importing it, to see if it would be smart enough to ignore the stuff that has already been changed, bit instead it duplicated all my and raptor's edits
11:00:23Watusimotoso let's try something else
11:01:22Watusimotowhat if you diffed your codebase and this version?
11:01:23Watusimotod846e3c285a0
11:01:39Watusimotothat's the latest of my edits you have in your codebase
11:02:50Watusimotoyou took that version, and merged it with your tree
11:03:26Watusimototherefore, it stands to reason that any differences between that version and your tree are the edits you've made
11:03:30Watusimotoif you look here:
11:03:31Watusimotohttps://code.google.com/r/bmartin9x-bitfighter/source/list
11:03:58Watusimotowhat we basically want is the sum of that left-most vertical line, which is all your edits
11:05:24martin_bedeokay, i've solved this
11:05:45martin_bedethe rdiff extension did just what we wanted the first time
11:07:09martin_bedeWatusimoto: but the diff is inverted
11:07:30martin_bedeis that a problem or is it just for reviewing?
11:08:40Watusimotono
11:08:57Watusimotothough you could perhaps get it "the right way"
11:08:57Watusimoto by reversing your diff parameters
11:09:23martin_bedethere is only one parameter, the url
11:09:27Watusimotobut I can deal with the backwards diff if need be
11:09:52martin_bedebut if i clone the official repo and then run this, i can get the normal cset
11:09:58martin_bedeit takes quite a while though
11:10:27Watusimotowould it be easir just to swap the +s and -s?
11:10:36WatusimotoI think that's all that would need to be done to invert it
11:11:20martin_bedei'm not sure that would result in a valid changeset
11:11:22martin_bedehttp://pastebin.com/HCqttYXe
11:11:26martin_bedehere is the inverted one
11:12:19martin_bedei'm cloning the official repo now, so the normal diff should be available later too
11:19:51martin_bedeWatusimoto: here's the not-inverted changeset: http://pastebin.com/Fc1RtEFN
11:20:17Watusimotogreat -- how did you do that?
11:20:46Watusimotoand what does this mean:
11:20:47Watusimoto/ I've got no idea why this works and getInt(L, 2) doesn't.
11:20:47Watusimoto+ //mSpawnTime = S32(getInt(L, 3) * 1000);
11:20:47Watusimoto+ //mTimer.reset(mSpawnTime);
11:20:58martin_bedewhat line is that?
11:21:04Watusimotoit's in core
11:21:12martin_bedei used the same extension
11:21:17Watusimotofrom here
11:21:18Watusimoto+++ b/zap/CoreGame.cpp Sun Jan 13 12:05:03 2013 +0100
11:21:34Watusimotoline 279
11:22:03martin_bedei pasted that there for refernce
11:22:04WatusimotoI'm trynig to understand if this is code that should be included but didn't work, or if it's just something you discarded
11:22:11Watusimotoso you could delete it?
11:22:13martin_bedeit's discarded
11:22:20martin_bedeyes, i'll remove it
11:22:23Watusimotook
11:22:24Watusimotogood
11:23:45martin_bedethere are two places where that comment is legit though
11:25:18Watusimotowhat do you mean by "legit"
11:25:38martin_bedei intended to left them there
11:26:30martin_bedebecause that may be a bug in other code
11:27:15WatusimotoI see you added @luafunc comments to document constructor params
11:27:17Watusimotothat's good
11:27:32Watusimotowould you please change them to @luaconst?
11:27:47martin_bedeokay
11:27:48Watusimotothat way our document generation can differentiate them from standard function sigs
11:27:59Watusimotobut good that you did that
11:30:48Watusimotoactually, have you made that @luaconst change yet?
11:30:55Watusimotobecaue I can do it easily in the diff
11:33:37martin_bedei did
11:34:21Watusimoto:-)
11:34:28Watusimotook, in asteroid, you have that comment again:
11:34:28Watusimoto/ I've got no idea why this works and getInt(L, 2) doesn't.
11:34:42martin_bedeyep
11:34:44Watusimotowhat does that mean?
11:34:57martin_bedelook at the next line
11:35:07Watusimotodoes it mean you expected to use L,2, but that failed
11:35:12martin_bedeyes
11:35:15Watusimotoand that L,3 worked for reasons you're not sure about?
11:35:18martin_bedeyes
11:35:29Watusimotoit's because geom occupied 2 slots
11:35:34Watusimotox,y
11:35:41Watusimotobut it does reveal a problem
11:35:51Watusimotoin that point(x,y) would only occupy one slot
11:36:00Watusimotoso that's something I need to figure out
11:36:30Watusimotobecause getInt(L, 3) wouldn't work if geom were just a point instead of 2 ints
11:36:46Watusimoto(or 2 nums)
11:37:36Watusimotook, I'll accept the code
11:37:42martin_bedeawesome
11:38:01WatusimotoYou've revealed a number of weaknesses that I had not considered before
11:38:11Watusimotoso I'm not terribly happy about it, but that's not your problem
11:38:28Watusimotoyou're going to do the other paramerization task?
11:38:30martin_bedemaybe later it will be :)
11:38:33martin_bedeyes
11:38:38Watusimotook
11:38:40Watusimotogood
11:38:44Watusimototake a look at this first
11:38:45Watusimoto+ // I've got no idea why this works and getInt(L, 2) doesn't XXXXX what if geom is a point, not an x,y pair?
11:38:45Watusimoto+ mSpawnTime = S32(getInt(L, 3) * 1000);
11:39:11Watusimotoif you are tempted to write that comment again, just make it L,2 for the moment, even if it doesn't work
11:39:18WatusimotoI've got to figure a way to make it work
11:39:29Watusimotobecause, as I said, it will fail
11:39:43martin_bedeokay
11:39:45Watusimotounless we require that the ints be in a table rather than just bare numbers
11:40:03Watusimotolocal s = Spawn.new({100,200}, 4)
11:40:14Watusimotothen the 4 would always be arg 2
11:40:31Watusimotoin fact, when doing your testing, test that way, and that will avoid the problem
11:40:49Watusimotothen getInt(L,2) will always work
11:41:16Watusimotoand I can decide whether to require that, or if there is a way to let (100,200,4) work
11:41:42Watusimotobut even as I'm writing this, I think that syntax should be eliminaetd
11:41:47Watusimotobecause it's very sloppy
11:42:06Watusimotoin fact...
11:42:57WatusimotoSpawn.new(point.new(100,200), 4) would have the best performance, as creating a point requires no object creation/garbage collection, whereas {100,200} does
11:43:44Watusimotosomething for me to ponder
11:44:05Watusimotobut, bottom line, don't test with Spawn.new(100,200,4) syntax
11:44:32Watusimotoonly with Spawn.new({100,200}, 4) or with points
11:44:36martin_bedealright
11:44:41Watusimotofingers are tired
11:45:11WatusimotoI'll accept your work, and apply your patch to the codebase
11:56:20Watusimotoyou should be good to go
12:13:42BFLogBot Commit: 3c5c35adc075 | Author: watusimoto | Message: Parameterize Core Lua constructor
12:13:44BFLogBot Commit: 609ff75480ce | Author: watusimoto | Message: Parameterize a few more lua objects
12:13:45BFLogBot Commit: d4ccfc5ed339 | Author: watusimoto | Message: Final part of Martin's paramaterization work
12:13:47BFLogBot Commit: 759a152f3a03 | Author: watusimoto | Message: Fix parameterization to work with points instead of raw coords
12:14:36Watusimotomartin, if you would, please move the @luaconst items after the @luaclass documentation item; I'll take care of the stuff that's already in, you just do it for your new code
12:14:54martin_bedeokay
12:15:02martin_bedewhat's that last commit?
12:15:26martin_bedei'm having a problem probably fixed by that
12:15:42Watusimotojust what we discussed -- using {100,200} instead of 100,200
12:16:13Watusimotoit removes those two comments... well, have a look, you'll see
12:16:46WatusimotoActually, pleaes put the @luaconst items up by the constructor itelf... that will make it easier to keep up to date
12:17:26Watusimotoas will be illustrated by my next commit
12:17:45BFLogBot Commit: 4e8db1703470 | Author: watusimoto | Message: Shuffle documentation around
12:26:12BFLogBot Commit: 9e9324157bfa | Author: watusimoto | Message: WARNING: THIS CHECKIN MIGHT BREAK SOME SCRIPTS!!! No longer allow list of bare numbers in place of a point -- this was bad practice and created problems with new parameterization code. Please specify actual points for geometry from here on out!
12:26:14BFLogBot Commit: 76d160a528cd | Author: watusimoto | Message: Typo
12:26:20Watusimotomartin_bede: this next checkin will affect your testing, but make your code easier to write
12:26:29Watusimoto(last checkin)
12:27:22martin_bedeokay
12:34:20martin_bedethis new thing disabled coordinate setting everywhere
12:34:40martin_bededespite that i use lines like this:
12:34:41martin_bedelocal a = Spawn.new({10,10});
12:35:10martin_bedeeverything has got a coordinate of 0,0 basically
12:36:21martin_bedeWatusimoto: is the line i posted here wrong or the new code?
12:45:26Watusimototry local a = Spawn.new(point.new(10,10))
12:45:47martin_bedethat results in invalid params error
12:45:48WatusimotoI'm trying to figure out if my latest change broke anything, and if so, what
12:45:57Watusimotook, hold on a minute
12:49:53Watusimotodid you do speedzone as part of your last batch of things?
12:50:10martin_bedeyes
12:50:22martin_bedewhy?
12:50:26Watusimotomaybe I botched the merge
12:50:31Watusimotothe L check was in the wrong place
12:50:38Watusimotoand the game crashed
12:53:29martin_bedeit's the same here too, strange. i remember editing it
12:54:26WatusimotoI've fixed it here, so it will get in the repo in a couple of minutes
12:56:16martin_bedeokay. anything about the 0,0 problem?
13:02:52Watusimotook, I see the error now
13:03:05Watusimoto(those other issues made things crash ebfore I got to my test level!)
13:03:33BFLogBot Commit: 28f35a8273f5 | Author: watusimoto | Message: Parameterization fixes: move L checks to outer block, make sure SpyBug runs intialize even if L is NULL
13:11:06Watusimotook, I see why it is crashing, in code, but why?
13:11:31mayanks43 Quit (Quit: Leaving.)
13:13:07martin_bedecrashing?
13:15:40Watusimotowell, not working
13:15:40BFLogBot Commit: f861b40434f5 | Author: watusimoto | Message: Fix geom checking code
13:15:45Watusimotook, the fix is in
13:15:49Watusimotothis code works for me now
13:15:50Watusimoto local f = ResourceItem.new(point.new(400,100))
13:15:50Watusimoto levelgen:addItem(f)
13:16:01Watusimotobut it didn't before that fix
13:19:33Watusimotomartin: where you specifially need a point in a signature (like for resource item), use PT instead of GEOM
13:19:52WatusimotoI think GEOM is becoming more trouble than it's worth
13:20:48martin_bedeokay
13:24:36WatusimotoI need to think about how to handle other geometries
13:27:16Watusimotomaybe for zones, PT3 for 3+ points? or POLY or ... something
13:30:39WatusimotoI'm heading out for a while in a few minutes, just so you know
13:31:37Watusimotomaybe PT, LINE, POLY for 1, 2, 3+ points
13:35:32martin_bedei'm almost ready with the second task. now changing GEOM to PT and moving luaconst lines
13:37:56martin_bedeoh didn't see you're going
13:38:27martin_bedethere are no items in the second task which require multiple points
13:39:03martin_bedewill you be able to review it before 6:00?
13:43:09Watusimotoyeah, the line/poly idea wouldn't happen until after you are done anyway... just musing to myself
13:43:18Watusimotomaybe before 6
13:43:26Watusimotohere it is 2:45
13:43:47Watusimotodo you need me to approve so you can grab another task?
13:44:11martin_bedeyes
13:44:31martin_bedei'm in Europe too
13:45:15Watusimotowell, I'll tell you what
13:46:01WatusimotoI'll take a look first thing when I get back.
13:46:17WatusimotoIf you have a good patch (like the last one) ready when I get back it will go quickly
13:46:26Watusimotoit probably won't be later than 6ish
13:46:27martin_bedeokay, thanks
13:46:42martin_bedei already submitted one just like that :D
13:46:46Watusimotoalready
13:46:48Watusimoto??
13:46:50martin_bedeyes
13:46:58martin_bedea few seconds ago
13:47:08martin_bedecset.diff
13:47:43martin_bedei've been working on this a while before actually claiming it, taking some risk
13:50:18LordDVG has joined
13:51:23Watusimototell you what
13:51:35WatusimotoI'll approve it now, but stick around this evening when I do a full review
13:51:41Watusimotothis will let you do something else this afternoon
13:51:47Watusimotoa quick look it looks fine
13:51:52Watusimotohow's that sound?
13:54:18martin_bedeawesome
13:55:12martin_bedemy next task will be the same too so i'll have to stick around no matter what
13:55:48martin_bedethere is a parameterizing one left
13:56:07Watusimotothere is?
13:56:11martin_bedeyes
13:56:12WatusimotoOh, I see a part 3
13:56:15martin_bede"group 2"
13:56:19WatusimotoI forgot whether I did 2 or 3 groups
13:56:29Watusimotolucky you... a couple of easy tasks!
13:56:48WatusimotoMarked as closed
13:56:58martin_bedeyea :D it took time to get up to speed though
13:57:01martin_bedethanks
13:57:35Watusimotolater
14:01:54martin_bedebye
14:03:10Watusimoto Quit (Ping timeout: 272 seconds)
14:17:57Watusimoto has joined
15:15:40LordDVG Quit (Remote host closed the connection)
15:27:13Darrel Quit (Quit: Leaving)
15:31:08Watusimoto Quit (Ping timeout: 260 seconds)
16:48:52Watusimoto has joined
18:24:11martin_bedeWatusimoto: hi. for now i can use GEOM for classes that require more than a point, right?
18:24:44Watusimoto Quit (Ping timeout: 272 seconds)
19:01:31Watusimoto has joined
19:19:39martin_bedeWatusimoto: hello. are you around?
19:20:02Watusimotohi
19:21:46YoshiSmb has joined
19:22:46YoshiSmbhi
19:23:21martin_bedehi YoshiSmb
19:24:49martin_bedeWatusimoto: how can i pass GEOM in the arguments? nothing seems to work
19:25:18Watusimotoyou should be able to; I didn;t disable it yet because I haven;t foud a good successor yet
19:25:38Watusimotowell, actually, I may have misunderstood
19:25:41Watusimotowhat do you want to do?
19:26:07Watusimoto(provide a short code snippet that illustrates, perhaps?)
19:30:51martin_bedehttp://pastebin.com/heXUhjNx
19:31:05martin_bedethese all fail
19:31:11martin_bedei tried them one by one
19:31:20martin_bedewhat's the right way to do that?
19:31:56Watusimotothe first should work
19:32:15Watusimotowhat error do you get?
19:32:39martin_bedeCould not validate params, expected geom, int
19:32:51Watusimototry this:
19:33:09Watusimotolocal aj = WallItem.new({point.new(100,100), point.new(111,111)}, 5);
19:33:27martin_bedewait, i sent the wrong error
19:34:13martin_bedeit expects point or geom
19:34:20martin_bedeby it i mean BfObject::setGeom()
19:34:42martin_bedeso not the constructor is throwing that, but a method it calls
19:36:18Watusimotook... do you have a stack trace?
19:37:23martin_bedethis is everything: http://pastebin.com/Fa7rwt9k
19:37:59raptor has joined
19:37:59ChanServ sets mode +o raptor
19:39:17martin_bede Quit (Remote host closed the connection)
19:40:13Watusimototry that line I posted above
19:40:21Watusimotolocal aj = WallItem.new({point.new(100,100), point.new(111,111)}, 5);
19:41:05Watusimotoalso, if you post your WallItem code, I can try to figure out what is going on here
19:41:10Watusimotohi raptor
19:41:32YoshiSmb Quit (Ping timeout: 245 seconds)
19:41:54martin_bede has joined
19:42:24YoshiSmb has joined
19:44:38martin_bedeWatusimoto: http://pastie.org/5679819
19:44:43martin_bedethat's the constructor
19:45:52raptorhi
19:45:55martin_bedehi
19:46:02raptormmmm... cheese + crackers lunch
19:46:24Watusimotowhy do you set the width before grabbing the geom?
19:47:20martin_bedebecause the value is lost after the pop
19:47:47Watusimotowhy do I already have this code?
19:47:48martin_bedeso i would have to create a variable
19:48:02martin_bedebecause it's not my current task
19:48:05raptorhey! dmitry claimed the mac notifier!
19:48:07martin_bedeprobably the first
19:48:21martin_bedeit was tested, and it worked
19:48:30Watusimotoraptor: !!
19:48:40Watusimotook
19:49:14Watusimotoand why do you need to pop it?
19:49:24Watusimotois that to clean things up for getGeom?
19:49:30martin_bedeyes
19:49:37Watusimotoraptor: poor donny mitsov...
19:49:45kaenhello all
19:49:55martin_bedehey
19:50:34kaenraptor, I got the world -> screen space function working. I can start the fun parts now :}
19:51:29kaenbut first I'll wrangle those cmake modules so we can get that put in.
19:51:34Watusimotoif you do something like that in other constructors, please try to comment why it is; I wrote this:
19:51:35Watusimoto setWidth(lua_tointeger(L, -1)); // Grab this before it gets popped
19:51:36Watusimoto lua_pop(L, 1); // Clean up stack for setGeom, which only expects points
19:51:52martin_bedeok
19:51:54raptorkaen: i can't wait!
19:52:14raptorhaha Watusimoto i like your first new comment to donny...
19:52:17Watusimotohi kaen
19:52:17raptoroh man
19:52:21kaenhi Watusimoto
19:52:26Watusimotoyes, well
19:52:32Watusimotoit sort of helped
19:52:49Watusimotobut he gave me a 1/2 meg diff for what I think is about 10 lines of code
19:53:00raptorwow...
19:53:02raptoruh
19:53:04Watusimotoand I'm not going to fish them out
19:53:10raptordo we know what system he is on?
19:53:21WatusimotoI've hand applied too many patches this week
19:53:25Watusimotolinux, I believe
19:53:31Watusimotolots of .i and .o files
19:54:00raptorbecause the command i kept telling him 'hg diff' *only* shows differences to commit files and should produce a nice small patch
19:54:31Watusimotowell he's doing something else
19:54:50WatusimotoI have some sympathy -- you know I have epic struggles with these tools myself
19:55:03Watusimotobut still
19:55:26Watusimotowell, martin_bede and I had a fight with hg, but we won in the end
19:55:43raptorpersistence!
19:55:56Watusimotothat is really the key to this whole endeavor
19:55:59kaensomething like "hg add" without any parameters will add all of the files in at least the current directory.
19:56:08kaenit's an easy mistake to make
19:56:14kaenI think it actually recurses...
19:56:17WatusimotoI hope that';s not what he did
19:56:21raptori hope that's not what he did
19:56:31kaenlol hive mind.
19:56:53raptoryou will be assimilated
19:57:03kaenI just hope that's not what he did
19:57:19kaen:P
19:59:15raptorok, the logs got huge again... and i don't wish to read them.. anything else I need to know?
19:59:39martin_bedeWatusimoto: did you find any clues yet?
20:00:02Watusimotowell, I was mostly chatting...
20:00:04Watusimotobut
20:00:07Watusimotothis adds a wall for me
20:00:08Watusimotolocal aj = WallItem.new({point.new(100,100), point.new(201,100)}, 20);
20:00:08Watusimotolevelgen:addItem(aj)
20:00:45raptora commit with a WARNING...
20:01:00Watusimotothis generated an error for me
20:01:01Watusimotolocal aj = WallItem.new(point.new(100,100), point.new(201,100), 20);
20:01:01Watusimotolevelgen:addItem(aj)
20:01:30Watusimotoso I think there is a bug in using multiple points, but enclosing them in a table looks good
20:01:58raptor gasps
20:02:05raptoryou edited lua-vec!
20:02:20Watusimotoraptor, kaen, sam686 (if you're there)... just so you know I've unilateraly depreicated and then removed the functionality of specifying "loose" coordinates instead of points in lua methods
20:02:41Watusimotothis used to work: t = TestItem.new(10,20)
20:02:45Watusimotonow it has to be
20:02:56Watusimotot = testItem.new(point.new(10,20))
20:02:59raptordeprecated, then removed == break everything?
20:03:09Watusimotonot sure if it breaks anything
20:03:15raptorbot rewrite time!
20:03:24WatusimotoI haven;t tested yet, but there aern't too many things that use this
20:03:31WatusimotoI doubt it affects bots at all
20:03:47Watusimotowhen I rewrote mazegen a whileback, I used the new syntax
20:03:49WatusimotoI think
20:03:51raptoroh yay your merge went smoothly
20:03:59martin_bedeWatusimoto: forgot to try that, enclosing them works here too
20:04:07Watusimoto:-)
20:09:04sam686I take that it won't affect my 6357.levelgen even if that levelgen doesn't ever use any of "Point
20:09:49sam686levelgen that only uses levelgen:addLevelLine is mostly unaffected..
20:11:27raptoraddLevelLine is the workaround to everything!
20:11:31raptorhi sam686
20:13:00sam686Oh, and what will happen when I do levelgen:addLevelLine("Barrier 50 0 0 1 1 1 0") in a middle of a game?
20:13:00Watusimotosam686: addLevelLine is completely unaffected, though I want to get rid of it a sometime in the future (not necessarily the near future)
20:13:33Watusimotosam686: untested, but should work
20:13:47WatusimotoBut better would be:
20:14:38Watusimotolevelgen:addItem(WallItem.new({point.new(0,0), point.new(1,1), point.new(1,0)}, 50))
20:15:16Watusimotocurrently you can, I think, add walls, but not modify or delete them
20:17:25raptorheh, donny's latest patch patches the hg store
20:17:44kaenlol
20:18:12YoshiSmb Quit (Ping timeout: 245 seconds)
20:19:13raptorand... i can't believe it... it's even harder to apply than the last one...
20:19:17Watusimototested this
20:19:18WatusimotoTimer:scheduleOnce(function() levelgen:addItem(WallItem.new({point.new(0,100), point.new(100,100), point.new(100,0)}, 50)) end, 3000)
20:19:39Watusimotoafter 3 seconds, a wall does indeed get added. and here's the best part... it's invisible!
20:19:47Watusimotobut it impedes the ship
20:19:50raptorhaha
20:20:37Watusimotoso raptor: there is how you can put a whole function into the timer
20:21:44raptorhaha
20:24:07kaenreally crazy that it's invisible.
20:24:15raptorserverside only!
20:24:23kaensince it actually does get added, and that function works in other contexts
20:24:41kaenmaybe the fact that it's lambda changes it scope?
20:25:02kaenI'm pretty sure lua has closure though.
20:25:30BFLogBot Commit: 1586a8600579 | Author: watusimoto | Message: Add (unused so far) try/catch block
20:25:31BFLogBot Commit: fab6764b7732 | Author: watusimoto | Message: Make test chnage from GEOM to PT. Seems to be fine.
20:25:33BFLogBot Commit: a4a093117f17 | Author: watusimoto | Message: Import latest paramterization changes from Martin -- still needs fuller review
20:25:34BFLogBot Commit: 018852911434 | Author: watusimoto | Message: Comment
20:25:36BFLogBot Commit: 33f696664184 | Author: watusimoto | Message: Add try/catch around startLua
20:25:37BFLogBot Commit: 4a8d9795f9b0 | Author: watusimoto | Message: Formatting
20:25:39BFLogBot Commit: 26270079b909 | Author: watusimoto | Message: Much cleaner error handling, more consistent lua error messages, reduced code, fewer passed values, and, well, just plain nicer framework for running lua scripts
20:25:57kaen\o/
20:26:17Watusimotosome of those things might be true
20:34:38martin_bedeWatusimoto: the third task is almost complete now, but ForceFieldProjector has got the same problem as Turret had
20:34:47martin_bedeit tries to position itself
20:34:58martin_bedeso 7 is added to x every time
20:35:21Watusimotoyes
20:35:31Watusimotomark it with a comment similar to what's in turret
20:35:35martin_bedeokay
20:35:51raptorLuaScriptRunner.cpp:270 and LuaScriptRunner.cpp:358:
20:35:58raptorerror: cannot pass objects of non-trivially-copyable type ‘const string {aka const class std::basic_string<char>}’ through ‘...’
20:36:04raptor:)
20:36:07raptoryay vc++!
20:36:41kaenme too
20:36:47kaenI just confirming it...
20:36:51kaenI was just*
20:38:14DonnyMitsov has joined
20:38:25Watusimotoah... what?
20:38:31Watusimotohi DonnyMitsov
20:39:04DonnyMitsovhi can you help me with my repo problems
20:39:08kaenfixed it.
20:39:14Watusimotoyes
20:39:30Watusimotocan you first explain how you made the most recent diff?
20:39:48Watusimotoand then we can see (maybe) what went wrong and how to fix it
20:40:09kaenhttp://pastie.org/5679998
20:40:15kaenraptor, can you verify?
20:40:54DonnyMitsovI cloned once my repo with the latest code and added the neccessary changes then i changed the name of the folder and cloned the repo again
20:41:45DonnyMitsovthen I used the command: diff -crB donimicov-task01 donimicov-task > donimicov-task.patch
20:42:04raptorhello
20:42:19raptorkaen: verifying... yeah, that fixed it..
20:42:56raptorkaen: you didn't add it to a clone?
20:43:03raptorif not, i'll jsut commit...
20:43:06kaenuh all of my clones are tied up
20:43:11kaenI'd have to make a new one
20:43:16kaenoh wait
20:43:16raptornahh
20:43:23kaenI have a bugfix clone
20:43:24kaenderp
20:43:25raptorwait - you have bugfixes branch?
20:43:29raptoryes that
20:43:32raptoryay
20:43:58kaenthat was a really good idea
20:44:04kaenso good I forgot I had it...
20:44:08raptorheh
20:44:29Watusimotoso DonnyMitsov, I'm thinking that one of your two repos has all these build files in them
20:44:41Watusimotoyou said you ran make clean; did you do that on both repos?
20:45:03Watusimotoraptor: make clean will remove all the build files, right? including .i files?
20:45:07DonnyMitsovno I did on the 2 previous
20:45:10raptorDonnyMitsov: instead of using diff, can you instead run 'hg diff' ?
20:45:15raptorwith 'hg' first
20:45:21raptorthat will make a nice small diff
20:45:44DonnyMitsovI made a patch from those 2 I just recently downloaded
20:46:23martin_bedeWatusimoto: did you test my last entry (the previous task)?
20:46:25Watusimototry hg diff
20:46:44Watusimotomartin_bede: not yet fully
20:46:50Watusimotoa little here and there
20:47:27BFLogBot Commit: ecf4c8b9d228 | Author: kaen | Message: Pass char* to string formatting function. Fixes build errors on gcc
20:47:28Watusimotoyou don't by chance have a test script that adds a bunch of these objects do you?
20:47:32sam686umm. Now I am getting ***LEVELGEN ERROR*** (null) on newest changes on hg..
20:47:52kaenooh a little quick raptor... I'm still building in my bugfix clone
20:48:02kaen:x
20:48:03raptoroops...
20:48:06Watusimotokaen: does that fix the issue raptor noted above?
20:48:08kaenI think it'll still work
20:48:11kaenWatusimoto, I think so
20:48:17raptori usually like to build *first* then commit... :)
20:48:21kaenheh
20:48:27martin_bedei don't have. my test just creates them and prints some of their values
20:48:44raptorkaen: no worries, it built
20:49:07kaenI want your processor :<
20:49:13kaenstill building here...
20:49:17raptori've got a number-cruncher
20:49:32raptori7 something
20:49:35kaenI've got a low-end consumer laptop
20:49:39kaen\o/
20:49:45raptorhehe
20:49:53raptorthe ones they sell by color!
20:50:00kaenI wish :<
20:50:04kaenmine's gray
20:50:18kaenI think I want to give it a paint job when my mobo comes
20:50:37raptori have a 3 yearold that is handy with water colors...
20:50:43kaenheh
20:51:05Fordcars_ has joined
20:51:08Fordcars_hi
20:51:12kaenhi Fordcars_
20:51:59Fordcars_quick question: for a dedicated server, is the command for bots -bots "totalbotsnumber"
20:53:26kaenI don't know if there's a command line parameter for it
20:53:30raptorFordcars_: i think you edit the bitfighter.ini instead
20:53:33kaenbut you should just set it in your bitfighter.ini
20:53:35kaenyes.
20:53:37martinb_android has joined
20:53:45Fordcars_ok because in zap you could
20:53:50Fordcars_thanks
20:53:51kaen; MaxBots - The max number of bots allowed on this server.
20:53:51kaen; BotsBalanceTeams - Enable bot auto-balancing in each level.
20:53:51kaen; MinBalancedPlayers - The minimum number of players ensured in each map. Bots will be added up to this number.
20:54:02DonnyMitsov Quit (Ping timeout: 245 seconds)
20:54:10raptordonny come back!
20:58:15BFLogBot Commit: e0f29332ab20 | Author: watusimoto | Message: Reduce code duplication
20:58:17BFLogBot Commit: 13882576fa63 | Author: watusimoto | Message: Formatting
20:58:18BFLogBot Commit: 8a2ab180977e | Author: watusimoto | Message: Is const better?
20:58:20BFLogBot Commit: 71b1333ab7d6 | Author: watusimoto | Message: Merge
21:02:15martin_bedebye
21:04:33martin_bede Quit (Quit: Bye)
21:06:53raptorbye!
21:28:24BFLogBot Commit: e0febeaf010c | Author: watusimoto | Message: Fix bug: restore ability to pass a list of points to a wallItem constructor (and others requiring multi-point geoms)
21:28:41Watusimotoso quick design question
21:28:55Watusimotocurrently, we specify the arguments a lua function can take like this:
21:28:56Watusimoto static LuaFunctionArgList constructorArgList = { {{ END }, { GEOM, INT, END }}, 2 };
21:29:21Watusimotoin this case, there are two options; either an empty list, or a GEOM followed by an INT
21:29:29WatusimotoINT is obvious what that is
21:29:37WatusimotoGEOM is a little more nebulous
21:30:25Watusimotoit was designed to be a flexible placeholder, meaning a Point for a point item like a TestItem, a point pair for something like a TextItem, or a whole slew of points for something like a LoadoutZone
21:30:41Watusimotofor various reasons, I now think this was a bad idea, and am looking for alterntives
21:30:56WatusimotoPOINT is the obvious replacement for testItem
21:31:10WatusimotoI'm thinking that LINE would mean two points
21:31:16Watusimotoand POLY would mean 3 or more points
21:31:33Watusimotoalternatives might be POINT, POINT2, POINT3+
21:31:51Watusimotowell, that's my only alternative at this point
21:32:06Watusimotowhat seems more intuitive when you read it?
21:32:16Watusimoto(or do you have no idea what the heck I'm talking about?)
21:41:43Watusimotowell, those are all good suggestions!
21:50:04raptorPOINT, ...
21:50:18raptorone signature to rull them all
21:50:25raptorrule
22:02:31kaenokay raptor, I've added all cmake modules that are included in the 2.8 package
22:02:53kaenwhich is all of the libs except modplug and vorbis
22:03:51kaenfor those two I've just used the switches from the Makefiles and crossed my fingers. It will work on at least every system it worked on before
22:04:34kaenI tested dedicated, master, and client builds, they all work. master still doesn't have config options though
22:05:45BFLogBot Commit: e5e44b878e64 | Author: sam8641 | Message: Faster getAngleDiff
22:13:40Watusimotoraptor: You may have misunderstood
22:14:04WatusimotoI meant I want to choose between POINT2 and LINE to mean a pair of points
22:14:13kaenLINE
22:14:14Watusimotoand between POINT3+ and POLY to mean 3 or more points
22:14:22kaenPOINT2 makes me think 2d point at first
22:14:33Watusimotothese are 2d points :-)
22:14:50kaenyes, but then I'd think POINT3 is a 3d point
22:15:06raptor Quit (Ping timeout: 248 seconds)
22:15:11Watusimoto3+d point
22:15:14Watusimotocould include time
22:15:22kaenmaybe it's just the warped state my mind is in from hanging around in OGL for the last two days, but that just seems potentially misleading
22:15:35WatusimotoI agree; I like LINE better myself
22:15:40kaenwell, a 4D point is just a 4D vector, which is a common graphics data structure.
22:15:47kaenbetter to be as unambiguous as possible
22:15:53Watusimotofully agree
22:27:42BFLogBot Commit: 218cec954146 | Author: sam8641 | Message: Fixed getAngleDiff - uses FloatPi instead of 360. Fixes ship's flame blink wrong way once on ship rotation
22:32:43Fordcars_haha I just got bored and made a Mac dedicated server app
22:33:28sam686Why is 127.0.0.1:28000 listed on server list? can't connect to it.
22:33:42sam686or 127.0.0.2 i meant
22:33:52Fordcars_happenned to me too before
22:34:09Fordcars_is it a test server?
22:34:13sam686haha, /connect bitfighter.org:28000
22:34:38sam686I guess the master picked the wrong IP address for Contest Server maps...
22:35:41kaenlol awesome
22:36:08Fordcars_lol
22:50:53YoshiSmb has joined
22:52:18BFLogBot Commit: f0b1868f6a5f | Author: watusimoto | Message: Documentation
22:52:19BFLogBot Commit: 7f0bfd14e513 | Author: watusimoto | Message: Remove some unused oglconsole code
22:52:21BFLogBot Commit: 5bb7a344d106 | Author: watusimoto | Message: Remove more unused code
22:52:22BFLogBot Commit: df6e12f1bad1 | Author: watusimoto | Message: Fix garbagy looking characters that I mistook for misrendered characters
22:52:24BFLogBot Commit: f68313672825 | Author: watusimoto | Message: Merge
22:53:43WatusimotoThe buglist is getting smaller, finally
23:03:10Fordcars_yay
23:04:01Fordcars_why is there always Google [Bot] showing up?
23:04:23Fordcars_"Google[Bot]"
23:09:52YoshiSmb Quit (Ping timeout: 245 seconds)
23:11:59raptor has joined
23:12:02ChanServ sets mode +o raptor
23:13:53raptorhi
23:13:55raptori'm back...
23:14:34raptoris that localhost address the contest server?
23:15:53raptorthat's better...
23:16:59kaenraptor, can you check the cmake stuff when you're free?
23:17:13kaenafaict it's good to go
23:17:49kaen(minus some stuff for the master)
23:21:30raptorok
23:22:28raptorlet me check it out...
23:23:59raptorkaen: which revison is the latest for your cmake stuff?
23:24:07raptori'm pulling from https://code.google.com/r/bkconrad-kaen-bitfighter/
23:24:14raptorwhich i think you have other stuff mixed in..
23:24:18kaenuh oh
23:24:24kaenhold on
23:24:31Nothing_Much has left
23:24:32kaenI might have made a bad push somewhere
23:25:27raptoror even, if you want to just zip up the files and send them...
23:26:28kaenwell, according to my hg diff it's exactly right...
23:26:40kaenoops
23:26:50kaenokay now I've pushed the rest I think...
23:27:14raptorthat was the correct clone?
23:27:34kaenoh gosh...
23:27:38kaenyes that clone was correct
23:28:01kaenI think I might have just pushed my shader stuff somewhere @_@
23:28:49raptordid you push to the correct one?
23:28:52kaenno I really think that clone is up to date
23:28:57kaenall should be well
23:28:59raptorok
23:29:15kaenalthough I just pushed my shaders into my local cmake clone...
23:29:26kaenso I'll have to reclone ._.
23:29:57kaenor maybe I can tortoise my way out of this
23:30:00raptorheh
23:30:03raptorhg strip!
23:31:31raptorok, just to verify, 76ef2ed4a931, is the latest commit on that one, correct?
23:33:47kaenyes
23:34:17raptorok
23:34:35raptorwhat are your basic building optoins?
23:34:59kaenCMAKE_BUILD_TYPE is the only one you should care about
23:35:04kaenthe rest are added by the find modules
23:35:16kaenyour options are Release and Debug
23:35:23kaen(case sensitive for all I know)
23:35:58raptorwhere are the module?
23:36:03raptor*modules?
23:36:04kaen./cmake/Modules
23:36:19raptorah ha!
23:36:20raptorthanks
23:36:24kaenyep
23:36:28kaendid it not find them?
23:36:44kaenyour killing me raptor. is it building for you?
23:36:49kaenyou're*
23:37:00raptori had old cruft lying around everywhere....
23:37:04raptori did an hg clean
23:37:13raptorok, now it's nice and organized...
23:37:54kaenalso, it makes C::B project files
23:37:57kaenthey work like a charm
23:38:01raptorreally??
23:38:04raptorthat's awesome!
23:38:07kaenyes
23:38:16kaenbut the generator makes absolute paths :/
23:38:27raptoryuk
23:38:45kaenbut all you need is cmake
23:39:10kaenand you don't need to actually build or anything to just get the c::b file
23:39:37kaencmake .. -G "CodeBlocks - Unix Makefiles"
23:39:45kaenif you want to try it
23:41:34raptorok, so... i have questions...
23:42:16raptor#1 i see you added the subdirectory 'master' in the main CMakeLists
23:42:24raptorhow did you make it NOT compile that?
23:42:30raptorwhen doing a normal release?
23:42:45kaenEXCLUDE_FROM_ALL
23:42:53kaenmake master
23:42:55kaenwill build it
23:42:59raptorahhh
23:43:03raptorok
23:43:10raptori see it in the master/CMakeLists..
23:43:21kaenyes
23:43:39kaenit's in mysql++ and the definition of bitfighterd as well
23:43:59kaenonly bitfighter and its deps are in the default target
23:44:24kaen(referring to executable/lib names)
23:44:40raptorok, next question..
23:46:35raptorhere are the dynamic library dependencies that you're binary requires that are extra from the stock 018 binary: http://pastie.org/5680569
23:46:55raptornow i can forgive the ogg/vorbis ones
23:47:16raptorbecause ALURE might be compiled without dynamic loading
23:47:24raptorso that's three, but the other 4?
23:47:36kaenwell, they all come from the find plugins
23:47:55kaenwhich fill variables, e.g. X11_LIBRARIES
23:47:59raptormaybe one or two have to do with your changes?
23:48:00kaenor OPENGL_LIBRARIES
23:48:09kaenmaybe all of them
23:48:21kaenglu is like from findopengl.cmake
23:48:26raptorlet me try extracting the cmake files and adding them to stock 018...
23:49:14kaensm looks like it's part of X11
23:49:55kaensame with ice
23:50:23raptoruuid, i *know* would mean another package to install on most distros
23:50:36raptorsorry if i'm a bit of a perfectionist..
23:50:41kaenno problem
23:50:51kaenbetter to hammer it out than find out about this a month from now
23:50:56kaenand have to relearn it all...
23:51:16kaenanyway I did something with boost that I was unsure of
23:51:32raptorwhat was that?
23:51:36kaenI brought in the find module for it because tnl wanted to use its headers
23:51:55raptorreally??
23:52:00raptorhuh - boost should not be part of tnl
23:52:11kaenlet me look again
23:52:19kaenit sounds like I misinterpreted something
23:53:54raptorneat! you can do 'make help'
23:54:14kaenwhat really?
23:54:17kaenlol I didn't know that
23:54:25raptoryeah and it lists the targets
23:54:35kaenalso I don't know what I was talking about with tnl and boost. it's bitfighter that's using boost
23:54:42raptorheh, yes
23:54:59kaenshould I be statically linking there?
23:55:08raptorwe don't need to inlcude boost libraries in anything because I made a conscious decision to use the header-only stuff
23:55:16kaenokay
23:55:23raptorand all needed headers are included in the tree
23:55:36kaenalright, so we don't really need to find it
23:55:37kaenbut
23:55:57kaencmake includes a module for it, and at the top gives an example for using header-only stuff
23:56:01kaenwhich I followed
23:56:11raptorok
23:56:14raptorthen that's probably fine
23:56:19raptori wonder where libuuid came from..
23:56:23kaenbut, whence the other lib?
23:56:25kaenyes.
23:59:53kaenokay, so the only files that include the string "uuid" (insensitively) are not touched by my cmake stuff

Index Search ←Prev date Next date→

These logs were automatically created by BFLogBot on irc.freenode.net.