irclog2html for #bzflag on 20040202

00:13.58*** join/#bzflag DTRemenak_ (~DTRemenak@host-225-173.dialup.innercite.com)
00:30.46*** part/#bzflag your_dog (~jake@216-24-111-62.access.naxs.com)
00:34.26*** join/#bzflag orchid (~bzgirl@toronto-hs-64-26-176-100.s-ip.magma.ca)
00:44.13*** part/#bzflag rembo (~rembo@seesink.kabel.utwente.nl)
00:59.44*** mode/#bzflag [+o MrApathyCream] by ChanServ
01:11.53*** join/#bzflag MrApathyCream (~MAC@balt-209-163-102-151.qis-dynamic-dialup.coretel.net)
01:13.20MrApathyCreamyo
01:13.29*** mode/#bzflag [+o MrApathyCream] by ChanServ
01:34.50*** join/#bzflag Carnage (strayer@dyn216-8-133-147.ADSL.mnsi.net)
01:41.47*** join/#bzflag DTRemenak (~DTRemenak@host-226-202.dialup.innercite.com)
01:54.22MrApathyCreamibot, seen Patlabor221
01:54.31patlabor221 is currently on #project24 (3h 52m 28s) #bzflag (3h 52m 28s).  Has said a total of 6 messages.  Is idling for 3h 49m 4s
01:57.51*** join/#bzflag your_dog (~jake@216-24-111-43.access.naxs.com)
02:01.41your_doganyone here?
02:01.51MrApathyCreamyup
02:02.21your_dogoh drat, you ;)
02:02.22your_dogj/k
02:02.25DTRemenakthere's usually someone here
02:03.58your_dogi was thinking of changing the name of the #bzmatches channel 2 #ctfleague, what do y'all think?
02:05.22your_dogi think i like the idea of leaving #bzflag for the developers, and then having a serperate channel for just talking on
02:06.45MrApathyCreamif it floats you boat
02:06.46your_dogno comments?
02:06.56DTRemenakI have no opinion :)
02:06.58your_dogheh
02:07.05your_dogi find that hard 2 believe :)
02:07.42DTRemenakthere are many things in life which are difficult to believe
02:08.10your_dogheh :P
02:09.07your_doglike how good looking i am ;)
02:10.44DTRemenakbeauty is in the eye of the beholder :P
02:12.32your_dog:P~~~~
02:13.03your_dogso what do y'all think?
02:13.10your_dogi know y'all think something, so tell me :)
02:14.24DTRemenakI think the channel topic is outdated :)
02:15.15your_dogok, i don't like your attitude :)
02:15.26triclops_pbis there a quick way to get the bounding box of a tank?
02:15.27DTRemenakheh
02:16.03*** topic/#bzflag by irc.freenode.net -> http://BZFlag.org/ || http://list.BZFlag.org/ || http://stats.BZFlag.org/ || http://bzbb.bzflag.org/ || http://cia.navi.cx/stats/project/BZFlag || ban exceptions /mode #BZFlag +e nick!*user@*.dip.t-dialin.net || Nightly RPM's http://files.noodleroni.com/pub/bzflag/nightlies/ || 1.10.4 released, maintenance on v1_10branch, HEAD is now 1.11 || http://bzflag.mebigfatguy.com/TriangleWorld.png || everybody wang-chung tonight
02:17.07DTRemenakin what context?
02:18.37triclops_pbi'm trying to make a fancy targeting box for the hud, but i need a neat way to actually determine the size of the box
02:19.54Patlabor221BaseLocalPlayer::bbox
02:20.05Patlabor221from MyTank()->
02:20.10DTRemenakyup
02:20.25triclops_pbhmm, that wouldn't work on RemotePlayer would it?
02:20.28Patlabor221no
02:20.38Patlabor221there is no bbox for a remote player
02:20.45Patlabor221the client dosn't do testing on them
02:20.59Patlabor221all tanks have the same bbox
02:21.03Patlabor221so you can use yours
02:21.12DTRemenakunless they have a diff. flag
02:21.18*** join/#bzflag blast007 (~cgiirc@66.62.140.111)
02:21.22triclops_pbyeah
02:21.34DTRemenakthe local player's bounding box is calculated in BaseLocalPlayer::update
02:22.00DTRemenakyou could apply similar code to the remote player
02:22.01triclops_pbheh, problems too if you have a particular flag
02:22.43triclops_pbor put it in Player and save the duplication
02:22.44Patlabor221best to move that to a generalised class
02:22.57Patlabor221and yeah put it in the common class
02:22.59DTRemenakyup
02:23.46triclops_pbwhere should i dump the code from update?
02:24.22triclops_pbupdateBoundingBox?
02:24.28DTRemenakI would think a Player::GetBoundingBox would be best
02:24.40DTRemenakthen reference that from LocalPlayer::update
02:24.41triclops_pbso it calcs it before it returns it
02:24.52DTRemenakand wherever else you need it for the remote player
02:25.49triclops_pbyup, coolies.
02:26.28your_dogDTRemenak: do you play much?
02:26.37DTRemenakyour_dog: nope
02:27.10your_dogdo you write a lot of the code?
02:27.52DTRemenakdepends how much is a lot
02:28.10DTRemenakthere are many other devs who contribute more & better code than I do
02:28.10your_dogdo you write any code? :)
02:28.14DTRemenakyes
02:28.18your_dogcool :)
02:29.11your_dogi don't think the Panthers are going 2 win the superbowl :(
02:32.06DTRemenakCIA is down?
02:33.51your_dogthis country might b in big trouble then :)
02:34.24scanlineDTRemenak: yeah, server is down.. either something went wrong, or this is the upgrade they had scheduled for february
02:34.31DTRemenakah, ok
02:40.23triclops_pbis there a reason why the bounding box gets the new velocity and applies it to the old position in an update, instead of just getting the new position?
02:41.46your_dogg2g, bed time 4 me :)
03:07.39*** join/#bzflag DTRemenak (~DTRemenak@host-224-73.dialup.innercite.com)
03:08.33triclops_pbDTRemenak: is there a reason why the bounding box calculation gets the new velocity and applies it to the old position in an update, instead of just getting the new position?
03:08.57DTRemenakprobably
03:09.19DTRemenakbut I don't know
03:09.51triclops_pbhmm, i wonder if anyone does...
03:13.58triclops_pbhmm, using seems to work without it
03:14.04triclops_pbs/using//
03:14.33DTRemenakheh...I doubt anyone knows
03:14.52DTRemenakthat very same code was in the first bzflag cvs checkin
03:15.57triclops_pbmust be legacy
03:15.59DTRemenakso whoever knew 4 years ago probably forgot :)
03:18.26MrApathyCream70%?
03:18.42DTRemenakheh
03:19.10DTRemenakdoes that mean 70% of 1.11 was in 1.7d6 or 70% of 1.7d6 remains in 1.11?
03:19.18MrApathyCreamthe later
03:19.23DTRemenakhmm
03:19.34DTRemenakeven after the UDP net changes and such?
03:19.44MrApathyCream*shrug* just a guess
03:19.48DTRemenakheh
03:20.43DTRemenakalmost
03:28.43Noodleman_laptophas anyone ever run sloccount on bzflag?
03:29.06MrApathyCreamgood game
03:30.06Patlabor221it's more then 70%
03:30.14Patlabor221most of the code is there, it's just moved around
03:31.12Patlabor221bzdb is probably the largest archetectural change
03:31.35Patlabor221but if you want to go back further, I have 1.7b2 somewhere
03:35.09*** join/#bzflag pyrobunny (~mauddib@cpe-024-165-180-044.midsouth.rr.com)
03:35.13pyrobunnyhello?
03:35.45creeperzhi
03:35.54creeperzuugh
03:35.58creeperzcarnage is pissing me off
03:36.14*** join/#bzflag pyrobunny (~mauddib@cpe-024-165-180-044.midsouth.rr.com)
03:36.20pyrobunnyok sorry... back
03:36.27pyrobunnyanyone in here?
03:36.51creeperzhi
03:36.58pyrobunnyhey creeperz
03:37.41pyrobunnywhere did all the ctf'erz go.....
03:40.04Noodleman_laptopi just ran sloccount on bzflag:
03:40.09Noodleman_laptopTotal Physical Source Lines of Code (SLOC)                = 77,792
03:40.09Noodleman_laptopDevelopment Effort Estimate, Person-Years (Person-Months) = 19.34 (232.11)
03:40.09Noodleman_laptop<PROTECTED>
03:40.09Noodleman_laptopSchedule Estimate, Years (Months)                         = 1.65 (19.81)
03:40.09Noodleman_laptop<PROTECTED>
03:40.11Noodleman_laptopEstimated Average Number of Developers (Effort/Schedule)  = 11.72
03:40.13Noodleman_laptopTotal Estimated Cost to Develop                           = $ 2,612,903
03:40.17Noodleman_laptop<PROTECTED>
03:40.59Patlabor221just a year and a half (*snicker)
03:41.11DTRemenakheh
03:41.17DTRemenakand 2.6 million
03:41.24Noodleman_laptophehe
03:41.27Noodleman_laptop:-)
03:41.31DTRemenaktime/money tradeoff
03:41.43Patlabor221one would hope you coudl get a better sourcebase for 2.6 mil and a year
03:42.24DTRemenakif you compressed bz's development time down to 1.6 years, I'd hate to see the turnover rate
03:44.13DTRemenakis there any quick way to see what cvs revision introduced a particular line of code?
03:44.39Patlabor221not that I know of
03:45.00MrApathyCreamDTRemenak: binary search :)
03:45.24DTRemenakI was hoping something requireing less effort :)
03:45.32*** part/#bzflag pyrobunny (~mauddib@cpe-024-165-180-044.midsouth.rr.com)
03:45.40Patlabor221you could perhaps search the changelogs for the line
03:45.48DTRemenaklike "cvs grep 'line of code'"
03:45.49Patlabor221if you can find all the change logs
03:45.53MrApathyCreamPatlabor221: so keep saying the world should be more than just objects that define where you don't go
03:46.06MrApathyCreamcould you explain to my little brain, what you envision?
03:46.24Patlabor221portal or tree based system
03:46.50MrApathyCreamisn't portals just a superset of what we are doing now?
03:46.56Patlabor221umm no
03:47.02Patlabor221there are no portals in BZ
03:47.06MrApathyCreami realize
03:47.12Patlabor221current system is a plane with objects
03:47.23MrApathyCreambut if you just slapped a bunch of other worlds connected....
03:47.26Patlabor221protal system you define the places you can go, the places you can't go, you texture
03:47.34Patlabor221that is a very crude portal system
03:47.51Patlabor221more of a hybrid
03:47.59MrApathyCreamso more like lo-level triangels?
03:48.03Patlabor221no
03:48.08Patlabor221lower level areas
03:48.18Patlabor221the spaces between the boxes and stuff become cells
03:48.32MrApathyCreamhow is that more than just terminology
03:48.33Patlabor221a system like the one UT uses would work well
03:48.59Patlabor221cus there can be more one triangle per cell
03:49.00creeperzhow many vertices does the high quality tank have?
03:49.02Patlabor221and you get CSG
03:49.07Patlabor221about 200ishI think
03:49.17creeperzhow many faces?
03:49.21Patlabor221it's hard to explain in text
03:49.26Patlabor221170ish I think
03:49.32creeperzi see
03:49.36Patlabor221gimme a sec and I can check
03:49.40Patlabor221if you need real counts
03:49.49creeperzi just needed an estimate
03:49.56Patlabor221it's really low
03:49.58Patlabor221why?
03:50.06creeperzjust wanted to know
03:50.39DTRemenakdoes anyone know where the MASSIVE_NASTY_COMMENT_BLOCK came from?
03:50.55creeperzi built a small tank model and its 100 verts and 199 tris
03:50.57MrApathyCream?
03:50.57Patlabor221MrApathyCream: any system that allows for a uniform surface defininion would be optimal
03:51.18Patlabor221creeperz: there is no need to keep any new models as low as the current tank
03:51.20DTRemenakMAC: localplayer.cxx, line 123
03:51.21creeperzcourse it looks more like a plane
03:51.23MrApathyCreamas one continuous system of triangles?
03:51.25Patlabor221the current tank is way small
03:51.27creeperz:-)
03:51.40creeperzi was just comparing my model to the bz tank model
03:51.52Patlabor221as something that can be broken into triangles eventualy, but it's nicer to have something higher level then just tris
03:51.54MrApathyCreamoh DTRemenak that is client side spawning
03:52.01Patlabor221something with topology and connection info
03:52.02DTRemenakahhhhh
03:52.07MrApathyCreamspline?
03:52.13Patlabor221that's one method
03:52.21Patlabor221quads, planes, surfs, there are many
03:52.40Patlabor221but as long as it's uniform, and you have one consisntet method do deal with all surfaces
03:52.51Patlabor221for collisions, lighing, everything
03:53.37Patlabor221like in a BSP, all surfaces are the leaves of the tree
03:53.57Patlabor221so the code just knows how to render one of them,just with a buttload of atributes
03:54.20Patlabor221so you don't have to have code that renders a box, and a pyramid, and a cylinder, etc.....
03:54.56Patlabor221keeping them as an ojbect dosn't really get us anything except easyer world creation by hand
03:55.24Patlabor221tho if that is a requirement of the maps ( that they be easy to hand edit ) then staying with an object system has many merrits
03:55.33Patlabor221but then also limits
03:55.41Patlabor221many more things need to be special cased
03:56.01MrApathyCreami'd really like to put that coldet stuff in
03:56.21Patlabor221then you should do so, I don't forsee the world changing to something else anytime soon
03:56.26MrApathyCreambut it seems that the drawing and collision stuff should be sepeaerted first
03:56.38Patlabor221that would be a good help
03:56.56MrApathyCreami can see tat it would be relatively simple to allow arbitrary convex objects
03:56.58Patlabor221I would like to see that so that I can replace objects with models
03:57.05Patlabor221but let them still collide as a box
03:57.05MrApathyCreamnot to say thats the end, just as a next step
03:57.34Patlabor221http://jeffm.bakadigital.com/images/notbox.png
03:59.42Patlabor221I have an idea for a portal system, it could be cool, but it's made more for places that would allow for the avatars to have tilt and roll
03:59.47creeperzpretty
04:00.20Patlabor221but it would work well for seamless exterior and interior scenes
04:00.57Patlabor221you could drive into a building and that building can have stuff inside
04:01.05Patlabor221or a cave, or tunnel
04:02.10MrApathyCream<still downloading>
04:02.25Patlabor221it's not that impresive
04:02.39Patlabor221just somethign that could go in the place of a box, and look difrent
04:03.04creeperzthatd look cool with BUrrow
04:03.21Patlabor221you don't burow into boxes
04:03.30creeperzyea
04:03.34creeperzof course :-)
04:03.48creeperzbut it would look cool wit hthe grass then like mud looking stuffs
04:04.40Patlabor221the texture actualy has alpha in it that dosn't show there, the grass hanges off in a fringe, not that hard edge
04:13.31*** join/#bzflag Patlabor221 (~JeffM@adsl-64-173-104-109.dsl.lsan03.pacbell.net)
04:13.32*** join/#bzflag shkoo (nils@wsip-68-14-214-95.ph.ph.cox.net) [NETSPLIT VICTIM]
04:13.32*** join/#bzflag triclops_pb (~caleb@203-219-160-62-nsw.tpgi.com.au) [NETSPLIT VICTIM]
04:13.32*** join/#bzflag [dmp] (~dennis@port151.ds1-aboes.adsl.cybercity.dk) [NETSPLIT VICTIM]
04:13.32*** join/#bzflag jacques (~foo@westcoast.activebuddy.com)
04:13.32*** join/#bzflag gonkulator (~brandon@c-24-9-213-236.client.comcast.net) [NETSPLIT VICTIM]
04:13.32*** mode/#bzflag [+o Patlabor221] by irc.freenode.net
04:21.36*** join/#bzflag Mozul (~kirk@ca-redbch-cuda1-c10a-a-255.stmnca.adelphia.net)
04:24.38triclops_pbhttp://www.cse.unsw.edu.au/~cbro903/bzflag/target_HUD.png
04:25.11Patlabor221fun
04:25.35Patlabor221tho would be cool to leave the old box there and just put the corners on the target box, so you know when you ahve the 2 lined up your on target
04:26.57triclops_pbyeah, when i get the box working properly i'll make it look a bit better and an arrow for when the target moves off screen
04:29.49triclops_pbi'll probably have it appear only on the person you've selected to hunt too
04:29.50creeperzhave the angle on the top still and altitude on the right?
04:30.14creeperzor just leave it just like that?
04:44.39*** join/#bzflag stibnite (is@aau.cx)
04:59.56MrApathyCreamhmm no CIA
05:00.05captain_protonyeah
05:00.07captain_protonthe server its on went down
05:00.15scanlineor something
05:02.46MrApathyCreamheh
05:02.50MrApathyCreamtagteam
05:06.31MrApathyCreamtriclops_pb: is the theory that's where you're gonna shoot?
05:12.12MrApathyCreamBTW: tanks can drive over bumps now
05:12.22Patlabor221ahh you added the jump?
05:12.24Patlabor221cool
05:12.28Patlabor221did you bump proto?
05:12.28MrApathyCreamyes
05:12.33MrApathyCreamno
05:12.37Patlabor221probably should
05:12.40MrApathyCreamit's not really a proto change
05:12.46Patlabor221well it's a logic change
05:12.50Patlabor221you don't want the 2 mixing
05:12.53Patlabor221or it's a cheat
05:13.12Patlabor221tho may not be important
05:13.25Patlabor221since it's not like there are many "old" 1.11 clients
05:13.47MrApathyCreamheh
05:29.43creeperzrenegade version?
05:32.36Patlabor221thats what I've been told
05:32.39Patlabor221that I'm making one
05:32.48Patlabor221I hope it is really cool
05:33.03triclops_pbMrApathyCream: no, its just used to highlight your present foe
05:33.27Patlabor221can you just do one with like "L"s at the corners?
05:33.34Patlabor221that are maybe 10% transperant?
05:33.39Patlabor221I think that would look neat
05:34.08triclops_pbyeah
05:34.17triclops_pbi'm tempted to use a texture though
05:34.45Patlabor221why?
05:34.54Patlabor221lot of fill on that
05:35.04triclops_pbi'm also tossing up as to whether or not the size of the box should be proportional
05:35.17triclops_pbwell its just an idea, but i probably wont do it
05:35.24Patlabor221probably, but clamped to a  min limit I would think
05:35.29Patlabor221a texture would let you do a lot more
05:35.48Patlabor221you could check for the existence of a texture, and if it's not there use lines
05:36.00Patlabor221the sofites will probably want the lines
05:36.07triclops_pbyep, that would work
05:36.29Patlabor221just call getTextureID, if it's -1 then the texture wasn't found
05:37.07triclops_pbahh, nice. How do i get it to load?
05:37.15Patlabor221if it is found it will load it
05:37.27Patlabor221when you want to use it, just call bind and pass in the ID
05:37.43triclops_pbso bz loads all the textures in the texture directory?
05:38.06Patlabor221no, it loads them if you call getTextureID and it finds the name you pass in
05:38.27Patlabor221so getTextureID("hud_target")
05:38.34Patlabor221if it return -1 then it wasn't found
05:38.41triclops_pbahh, thats neat
05:38.44Patlabor221if it returns anything else, then that is the ID it loaded it into
05:39.00Patlabor221so it loads them as you use them
05:40.11Patlabor221your in 1.11 right?
05:40.16triclops_pbyep
05:40.19Patlabor221ok
05:40.38triclops_pbnot the latest cvs, but i'll merge when closer to completion
05:40.47triclops_pbor whenever
05:40.47Patlabor221cus 1.10.4 didn't return an ID, it returned a texture object
05:41.24Patlabor221your fine then, I think I made the ID change like right after the tag for 1.10.4
05:42.39triclops_pbcool. now back to coding for money
05:42.49MrApathyCreamSo Patlabor221, client side Obstacle, and it's sub classes BoxBuilding, PyrBuilding etc,
05:43.06MrApathyCreamdo two things. At world creating time, gen Scene nodes (QuadWallSceneNode, etc)
05:43.21MrApathyCreamand hold pos/size/rot for later col detection handling
05:43.24Patlabor221yeah
05:43.52MrApathyCreamTo add coldet, i would think we would just remove pos/size/rot and replace with their Collision3d object.
05:43.59MrApathyCreamand do everything thru that
05:44.15MrApathyCreamone coldet object per Obstacle derived class. (since the scaling thing)
05:44.23MrApathyCreamdoes thies make senes?
05:44.42Patlabor221I wanted the world to have an object in a world database class, and that object had an obsicle that was just collisions and "state" info, and then also register a "drawing object" if there was a draw manager
05:44.52Patlabor221sure
05:45.08Patlabor221who has the list of obsacles?
05:45.15MrApathyCreamworld
05:45.17MrApathyCream.cxx
05:45.18Patlabor221k
05:45.38MrApathyCreamit seems that once the scene nodes are generated, there is no tie between Obstacle class and these nodes
05:45.49Patlabor221if it was me I'd just put a "world object" class in between them that has the collider ojbect and the drawing object
05:46.13Patlabor221no direct tie yeah, just the fact that the obsicle has the render nodes
05:47.05MrApathyCreamactually i thing the obstacle class doesn't hold onto the scene nodes
05:47.18MrApathyCreamit just gens them
05:47.19Patlabor221well that's good
05:47.28Patlabor221it shouldn't have anything to do with them
05:47.33MrApathyCreamno
05:47.43Patlabor221makes headless easyer
05:48.05MrApathyCreamso i would think we don't need specialized classes
05:48.12MrApathyCreamBox vs Pyr, etc
05:48.49Patlabor221not for obsacle
05:49.00MrApathyCreami guess this world object class would need to store drive thru shoot thru tho
05:49.08Patlabor221and not if you apply the same collision responces to all surface types
05:49.11MrApathyCreamalong with col object
05:49.22MrApathyCreamwhich we would, right?
05:49.27Patlabor221yeah, and any other atributes that we come up with
05:49.29Patlabor221like "base"
05:49.39Patlabor221or maybe later "spawn", or "hill"
05:49.48MrApathyCreammmm
05:49.56Patlabor221real game related stuff
05:51.09Patlabor221would be nice to just be able to give the world class some sort of generic collision volume, and say "smack it up"
05:51.29Patlabor221then all game code has to do is figure out how to respond to the collison
05:53.16MrApathyCreamyes
05:54.34MrApathyCreamOpenGLTexture gone yet?
05:54.40Patlabor221not quite
05:54.47Patlabor221it's geting nasty
05:59.36*** part/#bzflag Mozul (~kirk@ca-redbch-cuda1-c10a-a-255.stmnca.adelphia.net)
06:01.37Patlabor221so in XML/HTML, a tag that is just there to set some data, and dosn't really need a begin and an end would be <tag/ atribute='data'> ?
06:02.31Patlabor221or is it allways <tag>data</tag>
06:02.50scanline<tag attribute='data'/>
06:03.22SportChicklearner around?
06:03.26MrApathyCreamthe second
06:03.36MrApathyCreamoh, heh
06:03.39MrApathyCreamthat was scanline
06:03.40MrApathyCream<PROTECTED>
06:03.56MrApathyCream<tag/> == <tag></tag>
06:04.11Patlabor221ahh
06:04.35MrApathyCreambut <tag/> != <tag>  </tag>
06:04.45Patlabor221yeah I can see how they are diff
06:04.56Patlabor221when generaly do you put stuff between the tags?
06:05.17Patlabor221when you need output that gets the effects of the tagset?
06:05.17MrApathyCreamwell sub tags for sure
06:05.17Patlabor221yeah
06:05.22Patlabor221but I mean non tag data
06:05.47MrApathyCreamthe bigger the data the more likely it's PCData as opposed to attribute values
06:06.10Patlabor221so it's really up to the tag desgner?
06:06.21MrApathyCreamyou can't constrain the type of data between begina and end tags, so usually if you want that it's better to use attributes
06:06.24MrApathyCreamPatlabor221: yes
06:06.36usherm00f
06:06.40MrApathyCreamif you want validation, etc, use attributes
06:07.24Patlabor221what is better?
06:07.25Patlabor221<world size='300x400'>
06:07.25Patlabor221<!-- world stuff -->
06:07.25Patlabor221</world>
06:07.25Patlabor221or
06:07.29Patlabor221<world>
06:07.32MrApathyCream<world>
06:07.35Patlabor221<size>300x400</size>
06:07.37Patlabor221<!-- world stuff -->
06:07.39Patlabor221</world>
06:07.41MrApathyCream<PROTECTED>
06:07.43MrApathyCream</size>
06:07.46MrApathyCream</world>
06:07.47MrApathyCreamperhaps
06:07.57learnerhehe
06:08.00Patlabor221where would world stuff go?
06:08.04learnerw00t?  xml formatted worlds?
06:08.10MrApathyCreamwhat 'stuff'
06:08.11Patlabor221I am just trying to learn
06:08.18Patlabor221like the sky defininion
06:08.21Patlabor221and ojects
06:08.27MrApathyCream<world>
06:08.30MrApathyCream<PROTECTED>
06:08.34MrApathyCream<PROTECTED>
06:08.37MrApathyCream<PROTECTED>
06:08.37Patlabor221where is size?
06:08.38MrApathyCream</world
06:08.42MrApathyCreamsize is an attribute
06:08.46captain_protonMrApathyCream: yes, you can constrain the data between begin & end tags, but you need to use a schema to do it
06:08.48Patlabor221of world?
06:08.54MrApathyCreamcaptain_proton: yes
06:09.18MrApathyCreampat since size has three parts, i'd have a size element with three atts
06:09.53Patlabor221but would the ojbects go inside the size tag?
06:10.04MrApathyCreami wouldn't think
06:10.26MrApathyCreamsize is the size of the world, i take it
06:10.32Patlabor221yes
06:11.14MrApathyCream<world>
06:11.14MrApathyCream<PROTECTED>
06:11.14MrApathyCream<PROTECTED>
06:11.14MrApathyCream<PROTECTED>
06:11.14MrApathyCream<PROTECTED>
06:11.15MrApathyCream<PROTECTED>
06:11.17MrApathyCream<PROTECTED>
06:11.19MrApathyCream</world>
06:11.21MrApathyCreamor soemthing like that
06:11.55learnerthat's a bit funky
06:12.22learnerif it's a world param, then it can be made one.. <world xSize="100" ySize="100"> ... etc
06:12.36MrApathyCreamlearner: depends how many parms there are
06:12.41learnersure ;)
06:12.44MrApathyCreami'd agree if thats the only parms then what you say is right
06:13.29Patlabor221like this?
06:13.30Patlabor221e x="300" y="400"
06:13.32Patlabor221gahh
06:13.34Patlabor221one sec
06:13.44Patlabor221decent sized paste inc
06:13.45Patlabor221<world>
06:13.45Patlabor221<PROTECTED>
06:13.45Patlabor221<PROTECTED>
06:13.45Patlabor221<PROTECTED>
06:13.46Patlabor221<PROTECTED>
06:13.47Patlabor221<PROTECTED>
06:13.51Patlabor221<PROTECTED>
06:13.53Patlabor221<PROTECTED>
06:13.53MrApathyCreamyes
06:13.55Patlabor221<PROTECTED>
06:13.57Patlabor221<PROTECTED>
06:13.59Patlabor221<PROTECTED>
06:14.01Patlabor221<PROTECTED>
06:14.03Patlabor221</world>
06:14.19Patlabor221well it's missing thigns I see :)
06:14.20MrApathyCreami would prolly group the objects under their own node tho
06:14.34Patlabor221why?
06:14.59MrApathyCreamhomogeneous leaves i guess
06:15.06MrApathyCreamjust my preference
06:16.57MrApathyCreamthat way i can simply define xpath like /world/objects and no everthing under there is an object, rather than specifiying some filter
06:17.17Patlabor221objects and groups should be interchangeable
06:17.24Patlabor221if your going to go object based
06:18.10MrApathyCreamyup
06:19.12Patlabor221what else is there besided objects?
06:19.29MrApathyCreamflags?
06:19.38Patlabor221server spawed now
06:19.55MrApathyCreamworld weapons
06:19.56MrApathyCream?
06:19.59Patlabor221k
06:20.17MrApathyCreamflagheight which is poorly named and rarely if ever used
06:20.30MrApathyCream(meaning the how much z size is needed to allow a flag to spawn there
06:21.29Patlabor221http://jeffm.bakadigital.com/dono.xmap
06:22.11MrApathyCreamid's really shouldn't be numeric
06:22.16MrApathyCreamin xml land
06:22.29Patlabor221what should they be?
06:22.39MrApathyCreamthey need to start with a letter unfortunately
06:22.39scanlinenumeric ids are fine, they're all strings
06:22.47scanlinereally?
06:22.48MrApathyCreamscanline, i don't think so
06:22.53Patlabor221it's just a refrence
06:22.59MrApathyCreamnot if the attribute is of type 'id'
06:23.05Patlabor221odd
06:23.06MrApathyCreamand they must be a NMTOKEN
06:23.13MrApathyCreammeaning no spaces
06:23.33Patlabor221seems to be an odd arbitrary enforecement
06:23.40MrApathyCreamPatlabor221: it would be nice if you could name a group but not image it, then specify the use of it
06:23.43Patlabor221ok what if I use some other name instead of IS
06:23.43MrApathyCreamwith a transform
06:23.58Patlabor221that's what that does
06:23.59MrApathyCreamPatlabor221: agreed
06:24.20Patlabor221the group block dosn't add those ojbects to the scene
06:24.30MrApathyCreamit would be slick if 'desert' was an external entity
06:24.30Patlabor221not untill it's called in an object of group type
06:24.38Patlabor221it is
06:24.40MrApathyCreamthats cool
06:24.45Patlabor221it's the name of a skybox
06:24.45scanlineMrApathyCream: looks to me like nmchar can start with a digit
06:24.47MrApathyCreaman xml external entity
06:24.53MrApathyCreamNMTOKEN?
06:24.54scanlineer, nmtoken
06:25.00scanlinehttp://www.w3.org/TR/REC-xml#NT-Nmtoken
06:25.05scanlineNmtoken    ::=    (NameChar)+
06:25.10scanlineNameChar    ::=    Letter | Digit | '.' | '-' | '_' | ':' | CombiningChar | Extender
06:25.10MrApathyCreamall the validators i know choke on that
06:25.23MrApathyCream*shrug*
06:27.22MrApathyCreamPatlabor221: <object name="right buildging" type="group" id = "1">
06:27.22MrApathyCreams/b
06:27.29MrApathyCream<object name="right buildging" type="group" refid = "1">
06:27.33MrApathyCreamor soemthing like that
06:27.43Patlabor221why?
06:27.54MrApathyCream1 is the id of the group that was defined, no?
06:27.57Patlabor221yes
06:28.10Patlabor221woud you also use refid in the group def?
06:28.10MrApathyCreamright, so id is the id of THIS object, not the group
06:28.14MrApathyCreamneed separate atts
06:28.15Patlabor221ahh
06:28.22Patlabor221makes sense then
06:28.47Patlabor221tho no real need I can see to add IDs to placed objects
06:29.50learnerid's should generally be avoided if at all possible
06:29.53MrApathyCreamwell its nice as you get getElementById(id);
06:29.58MrApathyCreamfor tranasformations etc
06:30.05Patlabor221refname?
06:30.15Patlabor221<PROTECTED>
06:30.15MrApathyCreamthats fine, whatever
06:30.15Patlabor221<PROTECTED>
06:30.15Patlabor221<PROTECTED>
06:30.15Patlabor221<PROTECTED>
06:30.15Patlabor221<PROTECTED>
06:30.15Patlabor221<PROTECTED>
06:30.17Patlabor221<PROTECTED>
06:30.19Patlabor221<PROTECTED>
06:30.53*** join/#bzflag Dave2 (~dave@teppic.demon.co.uk)
06:42.17Patlabor221http://jeffm.bakadigital.com/other_dono.xmap
06:45.20MrApathyCream<image>data/grass> :)
06:45.33Patlabor221yeah reload, fixed that
06:46.18MrApathyCreamneed some </edge>
06:46.31MrApathyCreamand not <edge/>
06:46.59Patlabor221reload
06:47.37MrApathyCream<wallheight="10"/>
06:47.39MrApathyCreamneed att
06:47.46Patlabor221ahh
06:47.49Patlabor221hmmm
06:47.54Patlabor221z would probably work
06:48.21Patlabor221guess it should be part of the cell then
06:48.30*** join/#bzflag Tupone (~Tupone@ppp-82-84-123-24.cust-adsl.tiscali.it)
06:48.39MrApathyCreamperhaps
06:50.08Patlabor221reload
06:51.24MrApathyCreamscanline:
06:51.25MrApathyCreamError at file C:\/test.xml, line 3, char 15
06:51.25MrApathyCream<PROTECTED>
06:51.25MrApathyCream<PROTECTED>
06:51.31Patlabor221so validaton just gets you an easy way to make sure you have all closed tags and atributes right?
06:51.50scanlineMrApathyCream: interesting, but that doesn't appear to be what the spec says :)
06:51.56MrApathyCreamPatlabor221: valid
06:52.02MrApathyCreamheh
06:52.02scanlinePatlabor221: well, that's just well-formedness
06:52.11MrApathyCreami've never known a validator to allow it
06:52.16scanlinePatlabor221: read the spec, there's a difference between a document being well-formed and valid
06:52.22MrApathyCreamAnd i was sure NMTOKEN has to start with a letter
06:52.35MrApathyCreamyes
06:52.40MrApathyCreamnext you need a schema file
06:52.41MrApathyCream:)
06:52.43Patlabor221so a validator just make sure it's valid right?
06:52.50Patlabor221what does the schema do?
06:52.55MrApathyCreamvalidates
06:53.05MrApathyCreamwell formed means you got all the <> in the right spot, etc
06:53.11Patlabor221ahhh
06:53.15MrApathyCreamvalidate means that the document makes sense according to the rules you want
06:53.29MrApathyCreamie vert elements must have x y z atts and they must be numbers, etc
06:53.32Patlabor221like make sure pos dosn't have an attribute "bob"
06:53.35scanlinewell-formedness is something every XML document -must- have for it to count as XML and be parseable
06:53.37MrApathyCreamcorrect
06:53.46scanlinevalidity is application-specific and defined by a schema or DTD
06:54.03scanlinefor example, a <turkey/> in a bzflag world would still be well-formed, but probably not valid
06:54.06MrApathyCreamthere are validating parsers
06:54.18MrApathyCreammost actually
06:54.33Patlabor221how would a parsing lib give this to me?
06:54.33scanlinemost parsers people use in real-world apps don't validate
06:54.40Patlabor221how wouldI get the data into my structs?
06:54.45MrApathyCreamscanline: bah
06:54.46scanlineusually folks run the file through a separate program to validate it
06:54.54MrApathyCreamxerces validates and roxors
06:55.06MrApathyCreampat here's a simple xml file
06:55.14scanlineyeah, but xerces isn't nearly as common as expat, minidom, libxml, and such
06:55.18Patlabor221I have a sample XML
06:55.18MrApathyCream<test xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
06:55.19MrApathyCream<PROTECTED>
06:55.19MrApathyCream<obj id="5"/>
06:55.19MrApathyCream</test>
06:55.20scanlinePatlabor221: read about SAX and dOM
06:55.23MrApathyCreamand xsd
06:55.24scanlineer, DOM
06:55.50MrApathyCream<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">
06:55.50MrApathyCream<xsd:complexType name="TestType">
06:55.50MrApathyCream<xsd:sequence>
06:55.50MrApathyCream<xsd:element name="obj" type="ObjType"/>
06:55.50MrApathyCream</xsd:sequence>
06:55.51MrApathyCream</xsd:complexType>
06:55.53MrApathyCream<xsd:complexType name="ObjType">
06:55.53Patlabor221so it's that complex that it can just be told?
06:55.55MrApathyCream<xsd:attribute name="id" type="xsd:ID"/>
06:55.57MrApathyCream</xsd:complexType>
06:55.59MrApathyCream<xsd:element name="test" type="TestType"/>
06:56.01MrApathyCream</xsd:schema>
06:56.15scanlinePatlabor221: it's not that it's complex, it's that you should read the specs yourself...
06:56.17MrApathyCreamSax is an event driven api, DOM is an object model api
06:56.29Patlabor221I'm looking for overview now
06:56.30MrApathyCreamboth are pretty simple
06:56.32Patlabor221not implemting stuff
06:56.35scanlinePatlabor221: but if you want the reader's digest version... SAX lets you register callbacks to handle events, like tags opening, closing, data being read
06:56.44Patlabor221k
06:56.45scanlinePatlabor221: whereas DOM gives you a tree data structure that represents the XML document
06:57.07Patlabor221k
06:57.17scanlineIMHO the best thing to use for bzflag would be a DOM where all the world objects were subclasses of the generic element object
06:57.19Patlabor221don't know if the events would do me much good
06:57.35Patlabor221asuming there are still ojbects :)
06:57.37MrApathyCreamsax actually works well for most things
06:57.52MrApathyCreamunless you need to refer alot to history in the file
06:57.53scanlineso the entire bzflag world would always be represented as an XML document in memory, but you could look at the same data as specialized classes for each object, with draw() functions and important variables already extracted from the DOM
06:57.55Patlabor221I could see the event thing working well for like state machiens like GL
06:58.01scanlineit would be a lot like what jetstream did, but with XML instead of libdb :)
06:58.28scanlinePatlabor221: that's really not what SAX was designed for.. it's usually good either for stream-processing apps, or as a backend to build DOM on top of
06:58.31Patlabor221I found the ship you need in jetstream :)
06:58.53MrApathyCreamor huge xml files
06:59.23MrApathyCreamme wonders if there is a c++ variant of JAXB
06:59.35Patlabor221http://www.well-of-souls.com/homeworld/hws/hw2/images/display.htm?AKHiigaran_Interceptor.JPG
06:59.53scanlineheh.. yep, it's a ship
07:00.07Patlabor221and a bit curvy
07:00.21scanlinecurvy isn't necessarily good... take star wars episode I for example
07:00.27Patlabor221yeah
07:00.37Patlabor221this is just enough curvy I think
07:01.28Patlabor221where find DOM parser
07:01.42MrApathyCreamxml.apache.org
07:01.43MrApathyCreamxerces
07:02.44Patlabor221so if I wanted to use id=
07:02.52Patlabor221I'ld have to use like id="CELL_1" ?
07:03.06MrApathyCreamyes
07:03.21Patlabor221lame but workable
07:03.28MrApathyCreamas long as it starts with a letter and has no spaces
07:03.38*** join/#bzflag sean_wo (~s2d@213.146.171.11)
07:03.45Patlabor221this was writen by nixies wasn't it ? :)
07:03.54MrApathyCream*shrug* prolly
07:04.14MrApathyCreamunless your schema is really flat, navigating the DOM model stinks
07:05.12Patlabor221any reason why none of the dowload links work?
07:05.29Patlabor221ohh wait, this is the same thing you used in your GL thingy isn'tit?
07:05.53MrApathyCreamyes
07:06.08Patlabor221is scema a difrent file?
07:06.11MrApathyCreamyes
07:06.23Patlabor221got an example of one?
07:06.29Patlabor221not as a paste :)
07:06.32MrApathyCreamlook up in the history
07:06.33MrApathyCreamoh ok
07:06.39MrApathyCreamlemme look
07:08.54Patlabor221was there one in your gl sample?
07:09.03MrApathyCreamno
07:09.12Patlabor221so they are not needed, just nice to have?
07:09.17MrApathyCreamyes
07:09.43Patlabor221and you did it with the event thign?
07:09.53Patlabor221in your gelly sample?
07:10.01MrApathyCreamhttp://bzflag.mebigfatguy.com/window.xsd
07:10.08MrApathyCreamum
07:10.14MrApathyCreamprolly sax, letsee
07:10.51MrApathyCreamyes, sax
07:10.56MrApathyCreamSAX is much simpler todo
07:11.55Patlabor221*shudder* ugg glut
07:12.42MrApathyCreamxsd goes with http://bzflag.mebigfatguy.com/window.xml
07:13.13MrApathyCreambut sax requires you to have your own storage of data
07:13.21Patlabor221that is what I want
07:13.36Patlabor221cus I'm gonna get tricky with the shizzy
07:13.42MrApathyCreamw00t
07:13.54Patlabor221ohh but not in this project
07:14.06Patlabor221not yet
07:15.10MrApathyCreamthe key is to derive your class from DefaultHandler
07:15.16MrApathyCreamand just override the events you want
07:16.28Patlabor221well it has merit for object based worlds. don't know if I like it for models, would think a standard format that modelers support would be best for that
07:17.12Patlabor221I don't know if surface based worlds will be simple enough to hand edit
07:17.16Patlabor221is there a binary XML?
07:17.23captain_protonyeah
07:17.24captain_protonEBML
07:17.37Patlabor221it save much space?
07:17.45captain_protoni think so
07:17.55MrApathyCreamPatlabor221: you can also write XMLReaders that convert arbitrary data into SAX events, so that you can plug and play
07:18.06Patlabor221huh?
07:18.50MrApathyCreamif you build your code based on SAX for instance, you can write an XMLReader that reads an Obj format file, and outputs SAX events to your code
07:18.51captain_protonhttp://ebml.sourceforge.net/
07:19.07Patlabor221ahhh
07:19.40Patlabor221generaly objs would just be refrences in the XML but it could be nice to use the same internal interface
07:19.59Patlabor221can it read binary too?
07:20.36MrApathyCreamattribute values can be arbitraty data
07:20.46MrApathyCreamif defined as entities
07:20.55Patlabor221cus MD3 is a pure binary format
07:20.57MrApathyCreamand can reference external sources
07:33.55Patlabor221so the schema for say the world would be like this
07:33.56Patlabor221<xsd:complexType name="world">
07:33.56Patlabor221<xsd:sequence>
07:33.56Patlabor221<xsd:element name="sky" type="skyType" minOccurs="0" maxOccurs="1"/>
07:33.56Patlabor221<xsd:element name="cell" type="cellType" minOccurs="1"/>
07:33.56Patlabor221</xsd:sequence>
07:33.58Patlabor221</xsd:complexType>
07:34.00Patlabor221?
07:36.16MrApathyCreamyes
07:36.31Patlabor221what is the xsd: for a float?
07:36.41MrApathyCreamits a prefix that specifies a namespace
07:36.47Patlabor221I know
07:36.48MrApathyCreamthe namespace defines a set of vocabulary
07:36.54captain_protonoh how i wish schemas had a more elegant way to just say "any of these, in any order"
07:36.55Patlabor221but where is a list of the words?
07:37.07MrApathyCreamelement, attribute, etc
07:37.10Patlabor221order is implied?
07:37.11MrApathyCreamin the case of xsd
07:37.18Patlabor221I see xsd:integer
07:37.19MrApathyCreamwith sequence yes
07:37.22Patlabor221what do I use for a string
07:37.24Patlabor221and for a float
07:37.27captain_protonPatlabor221: order is required in a <xsd:sequence>
07:37.29MrApathyCreamxsd:string
07:37.35Patlabor221eww that sucks, why does it need order?
07:37.44MrApathyCreamyou don't need to use sequence
07:37.53MrApathyCreamxsd:any
07:37.55MrApathyCreamxsd:all
07:37.58MrApathyCreametc
07:38.06MrApathyCreambut it is usual that you use sequence
07:38.14MrApathyCreamxsd:choice
07:38.40Patlabor221so xsd:choice instead of the seq?
07:38.45captain_protonaha, a complex type containing <xsd:choice maxOccurs="unbounded">
07:38.49MrApathyCreamyes youc an
07:38.54captain_protonwhat a hack :P
07:39.04Patlabor221what about for a floating point number?
07:39.05MrApathyCreamcaptain_proton: yup
07:39.17MrApathyCreamxsd:float i believe
07:39.19MrApathyCreamlessee
07:40.29MrApathyCreamxs:decimal
07:40.59Patlabor221why does int have a xsd not xs?
07:41.14MrApathyCreamwell just pick one and be consistent
07:41.17MrApathyCreamthe prefix can be anything
07:41.24Patlabor221ohh
07:41.25MrApathyCreamat the top of the file you specify what the prefix maps too
07:41.39Patlabor221xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"
07:41.41MrApathyCreambut typically schema files it's xsd
07:41.41Patlabor221that part?
07:41.43MrApathyCreamyes
07:44.32triclops_pbhttp://www.cse.unsw.edu.au/~cbro903/bzflag/target_HUD3.png
07:44.38Patlabor221http://jeffm.bakadigital.com/xmap_schema.xsd
07:44.50Patlabor221ooo looks cool
07:45.04triclops_pb:)
07:45.13MrApathyCreamchoice only allows one, is that what you want?
07:45.22Patlabor221ohh no I want any
07:45.32Patlabor221in any order
07:45.38MrApathyCreamattributes must be listed after the element block
07:45.45Patlabor221odd but ok
07:45.50MrApathyCreamso after the choice
07:47.32Patlabor221it dosn't like xsd:any
07:48.12Patlabor221vc.net will show the schema as a tree :)
07:48.53MrApathyCreamthe attributes in color type should have use="optional"
07:49.08MrApathyCreamsince you have red and r
07:49.13Patlabor221k
07:49.20MrApathyCreamand perhaps a default="1"
07:49.22Patlabor221same for X Y and Z
07:49.43MrApathyCreamas long as you have a default
07:50.25MrApathyCreamok, you've defined all the types, now as the last thing before the </xsd:schema> you need to specifiy what your document is
07:50.27MrApathyCreamlike
07:50.37Patlabor221?
07:50.48MrApathyCream<xsd:element name="bzw" type="world"/>
07:51.01MrApathyCreamthis says the root element of your world is bzw
07:51.04MrApathyCreamand is of type world
07:51.14MrApathyCream(for instance)
07:51.15Patlabor221ahh
07:52.09MrApathyCreamwhen your done, let me know and i'll validate them
07:52.29Patlabor221it didn't like xsd:any
07:52.35Patlabor221only all, seq, and choice
07:53.22MrApathyCreamodd
07:53.33MrApathyCreamyou are validating? cool
07:54.09Patlabor221.net underlines things it dosn't like
07:54.11MrApathyCream<xsd:any minOccurs="0"/>
07:54.14Patlabor221and has a scement viewer
07:54.15MrApathyCreamah ok
07:54.33MrApathyCreamany means an element not defined by the schema
07:55.04Patlabor221can I just but elements in the base of the complex type?
07:55.13Patlabor221or do they have to be in a sequence or choice?
07:55.23MrApathyCreamthey must be in some grouping
07:56.00Patlabor221is there a grouping that allows for any order?
07:56.49MrApathyCreamwell, as cappy says you can do choice, and then the element that is of that type can have maxOccurs="unbounded"
07:56.59Patlabor221ahh
07:58.06Patlabor221and what type do I use for id?
07:58.20MrApathyCreamxsd:id
07:58.38MrApathyCreamand xsd:refid for the one that refers to something else
07:59.45MrApathyCreamobviously id attributes must be unique within the file (validation rule)
08:02.37Patlabor221what do I put for data inside of a <tag></tag> set?
08:03.12Patlabor221or should my <imgage>file</image> really be <image ref="file"/>
08:04.02MrApathyCreamthe fact that it's complex does that
08:04.21Patlabor221what is right tho?
08:05.25MrApathyCreamif file is binary it should be an attribute
08:06.05MrApathyCreamif you are just using a file name, it really doesn't matter
08:06.46MrApathyCreamYou can specifiy that the attribute is an external entity that refers to a file, and the parser will read it while parsing
08:06.57Patlabor221don't want it to
08:07.02MrApathyCreamif you do this yourself as a separate step it doesn't atter
08:07.04Patlabor221it's not a file
08:07.08MrApathyCreamk
08:07.10Patlabor221it's a name of a texture entity
08:07.17MrApathyCreamok, then it doesn't matter
08:07.26MrApathyCreamattributes are always easier to handle while parsing tho
08:08.54*** join/#bzflag cniV (~kchan@c211-30-113-108.carlnfd2.nsw.optusnet.com.au)
08:14.46Patlabor221http://jeffm.bakadigital.com/other_dono.xmap
08:14.54Patlabor221http://jeffm.bakadigital.com/xmap_schema.xsd
08:16.13MrApathyCreamtypically you would tie the two together by adding xsi:noNamespaceLocation attribute in the xml file that points to the xsd file
08:17.15Patlabor221what if one doesn't know where that file is
08:17.22Patlabor221also how would I do an enumeration?
08:17.55captain_proton<xsd:restriction base="type"><xsd enumeration value="blah"/> ... </xsd:enumeration>
08:19.14MrApathyCreamyup
08:19.23MrApathyCreamadd multiple enumeation elements
08:19.53MrApathyCreamPatlabor221: the files is usually a url
08:20.04MrApathyCreambut if you don't know.... heh
08:20.06MrApathyCreamsol
08:20.17Patlabor221<PROTECTED>
08:20.17Patlabor221<PROTECTED>
08:20.17Patlabor221<PROTECTED>
08:20.17Patlabor221<PROTECTED>
08:20.17Patlabor221<PROTECTED>
08:20.18Patlabor221<PROTECTED>
08:20.22Patlabor221<PROTECTED>
08:20.24Patlabor221<PROTECTED>
08:20.34MrApathyCreambase is probably xsd:string
08:20.42captain_protonyup
08:21.07Patlabor221re-uped file
08:21.39Patlabor221and yes cap it is a verb when you use Myers/Fugikawa language compression
08:21.43Patlabor221:)
08:21.50captain_protonbah!
08:21.59Patlabor221you just need new codecs
08:22.26MrApathyCreampat your xml root is world, but the schema says it should be the_world
08:22.57Patlabor221so,  <xsd:element name="world" type="world" />
08:22.57Patlabor221<PROTECTED>
08:23.02MrApathyCreamyes
08:23.04Patlabor221k
08:23.09MrApathyCreamnot sure if name collision will occur
08:23.16Patlabor221reloaded
08:23.25Patlabor221call it worldType?
08:23.26Patlabor221k
08:23.44Patlabor221done
08:24.19Patlabor221I should make the image thing be an atribure huh?
08:25.05Patlabor221<image ref="data/dirt"/>
08:26.40MrApathyCreamdoesn't have to be but will be easier
08:26.45MrApathyCreamI don't think you can do this
08:26.46MrApathyCream<xsd:choice maxOccurs="unbounded">
08:26.55MrApathyCreammaxOccurs can only be put on the element
08:27.00MrApathyCreamiirc
08:27.20Patlabor221then how do I make it do any?
08:28.31MrApathyCreamremove this       </xsd:enumeration>
08:28.34MrApathyCreamoh
08:29.00Patlabor221updated
08:29.31MrApathyCreami'm not 100% certain that you can't but if you can't you'd have to add an element of type 'ThisorThat' and define that element to maxOccurs="unbounded'
08:29.38MrApathyCreamusually strict ordering is understood
08:29.56Patlabor221so people just live with ordering?
08:30.29MrApathyCreamyes
08:30.30Patlabor221I want to only have to put in the items that are difrent from the defaults
08:30.31MrApathyCreamheh
08:30.34MrApathyCreamthis isn't c++
08:30.39MrApathyCreamxsd::string s/h xsd:string
08:31.15Patlabor221fixed
08:32.01captain_protonhehe
08:34.27MrApathyCreamoptinal
08:34.30MrApathyCreamspelled optional
08:36.07Patlabor221ohh did I miss one?
08:37.53MrApathyCreamvector type x y z should be attributes, no?
08:38.49MrApathyCreamoops i meant IDREF, not refid
08:39.32MrApathyCreamxsd line 71 optinal
08:41.35Patlabor221fixed
08:43.26Patlabor221fixed the pos thing too
08:43.48MrApathyCreamIDs can't have default values, since they must be unique
08:43.58Patlabor221ahhh
08:44.00MrApathyCreamline 71
08:44.35Patlabor221fixed
08:45.27MrApathyCreamdumb rule: ID and IDREF must be caps
08:46.05Patlabor221<PROTECTED>
08:46.07Patlabor221like that?
08:46.21MrApathyCreamthe name value can be anything
08:46.25MrApathyCreambut yes the type is correct now
08:46.34Patlabor221k
08:48.09Patlabor221thanks for the info
08:48.17Patlabor221I'll look at it more tomorow
08:48.28Patlabor221if I survive the day sane
08:48.40MrApathyCreamheh imageType mandates that it has subelements but none exist in xml
08:48.41MrApathyCreaml8r
08:49.03Patlabor221it does what?
08:49.24MrApathyCreamyou say that imageType has uvScale, uvShift, uvRotate, but the xml doesn't have them
08:49.43Patlabor221ohh yeah, that file uses the defautls
08:49.50Patlabor2211,1,1
08:49.52MrApathyCreamthen say minOccurs = 0
08:49.57Patlabor221ahh ok
08:49.59MrApathyCreamfor uvScale
08:50.00MrApathyCreametc
08:50.17Patlabor221so min is 1
08:50.18Patlabor221not 0
08:50.25MrApathyCreamthe default is 1,1
08:51.09Patlabor221updated
08:51.27Patlabor221now I sleep :)
08:52.00MrApathyCreamheh, tim isn't gonna like pats checkin
09:26.01*** join/#bzflag triclops_pb (~caleb@203-219-160-62-nsw.tpgi.com.au)
09:43.17*** join/#bzflag larsl (~larsl@c213-100-50-18.swipnet.se)
09:43.17*** mode/#bzflag [+o larsl] by ChanServ
09:43.43larslShots are moving up and down now if you have a vertical velocity when you fire - is that a bug or did someone actually change that?
09:44.14larslIt will make it very hard to hit people on low buildings far away.
09:44.58triclops_pbi think its a new feature
09:45.23triclops_pbMAC and Pat were talking about it before
09:45.30larslEven the thief laser doesn't go straight. =P
10:07.36shkooi really like having vertical velocity.  it makes a lot more sense.
10:07.49shkooif it gets changed back it should at least be a configurable variable
10:12.30larslA server option would be OK with me.
10:28.37larslCool, are we using ALSA now?
10:33.11triclops_pbnot sure, check the platform code. there's sdl stuff in there.
10:36.50larslI just saw that `ldd bzflag` lists libasound.
10:37.37triclops_pbhmm, i dunno. ask tupone when he comes online he might know.
10:38.55*** join/#bzflag [dmp] (~dennis@port151.ds1-aboes.adsl.cybercity.dk)
10:40.21larslHello, [dmp].
10:40.41[dmp]Hi Lars
10:40.56larslFound any good algebra books?
10:41.42[dmp]Nope. Have to buy books for this semester. 1400 dkkr .. thats almost half of my income this month. And im still suffering from xmas/new year
10:42.13[dmp]607 for one book. Ill never understand why its so damn expensive
10:42.37[dmp]But this semester is gonna be interesting. Theme is compilers and languages.
10:44.36[dmp]We got: Syntax and semantics, Compilers and languages and Real time systems.
10:44.42larslBut that's as far as I'm going in that direction. No compiler construction etc, although it would be interesting.
10:45.19[dmp]:)
10:45.29[dmp]Yeah, the pay sucks tho
10:46.24[dmp]could ofcause create a bzlanguage. n "hello world" :-)
10:46.32larslWould you get paid to go to school for 20 years? We only get 6 years here.
10:48.38[dmp]No, i wont. We got 70 months.
10:49.45[dmp]And ive used 30-35 of them.
10:49.52larslCool. That's still more than in Sweden.
10:50.01[dmp]So i've got plenty. And i dont think im gonna need them.
10:50.32larslWe have some number of weeks that basically translate into 6 years of studies, if you do something else during the summers.
10:52.20larsl240 weeks.
10:52.52[dmp]Ah, yes. We can do that too. Decline our pay and store them for later. Never used it myself, but know that some do for sommerjobs etc
10:57.28larsl~studylarsl
10:57.31larsl, you lazy bastard! You should be studying right now!
10:57.35larslYes, ibot.
11:53.10*** join/#bzflag your_dog (~jake@216-24-110-11.access.naxs.com)
11:53.30your_doggood morning
11:59.47*** join/#bzflag triclops (~caleb@203-219-160-62-nsw.tpgi.com.au)
12:31.17your_doghi chestal
12:38.21*** join/#bzflag rooter ([eiI6rHX3n@webguy.cc.westga.edu)
13:38.57*** join/#bzflag drinian (~drinian@pcp03995358pcs.hamntn01.nj.comcast.net)
14:00.58*** join/#bzflag brlcad (~brlcad@chr160dhcp1509.chrchv01.md.comcast.net)
14:00.59*** mode/#bzflag [+o brlcad] by ChanServ
14:37.57*** join/#bzflag [dmp] (~dennis@port151.ds1-aboes.adsl.cybercity.dk)
14:43.29*** join/#bzflag mmu_man (~revol@ALyon-109-1-1-244.w217-128.abo.wanadoo.fr)
14:47.58*** join/#bzflag CIA-4 (micah@hyperreal.info)