#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2014-10-17

Timestamps are in GMT/BST.

00:01:17Invisible has joined
00:02:18sam686I redirected my internet's tcp 22 into my virtual machine tcp port 2222, and I got this: http://sam6.25u.com/upload/text1410/141016_22-08-26.txt
00:02:34sam686linux probably would have gotten hacked if those commands did run...
00:02:48raptorwhoa!
00:03:41raptorthat's really cool
00:03:43raptor!
00:03:49raptoryou set up a honeypot
00:05:32sam686I just use the pre-built honeydrive image, but use the latest git version of kippo, the good thing about kippo as I just run start.sh, theres no manual compiling as its python
00:08:48raptorthat's really neat!
00:09:13raptorlooks like the same bot struck several times - tried to pull junk from a frade8c domain
00:10:24sam686it looks like compiled binary file to me when I downloaded into windws (windows can't run linux binary)
00:12:04fordcarsNight!
00:12:07fordcars Quit (Quit: Page closed)
00:15:43Invisible Quit (Ping timeout: 272 seconds)
00:16:19raptori found my c++ exception problem: see the second bullet item here: http://luajit.org/extensions.html#exceptions
00:16:44raptorbut i'm not sure what to do about it because luajit can't handle throwing exceptions in Lua frames on all platforms
00:22:39sam686i wonder why x64 does it and not x86 on windows?
00:23:13raptorI think I'm going to need to hunt down and remove all of our 'throw LuaException...' and replace them with something else
00:31:58raptorOh good, I think I found a way...
00:42:30sam686Robot.cpp line 45/46, in Robot::Robot (Lua) constructor, that looks like one of a few places it uses a faster checkArgList(L,functionArgList,char,char), avoiding a slow string comparison of checkArgList(L,functionInfos,char,char)
00:44:45raptorI didn't know there were two options
00:47:00sam686although by the time it gets to checkArgList(L,functionArgList,char *,className,char *functionName), both the "className" "functionName" is only used for throw LuaException
00:48:04raptoroh no! my visual studio 2013 express license expired!
00:51:17sam686i wonder why do the express version always expires when offline for more then a month? 2008/2010 express never expire once activated though email..
00:51:55sam686well, newer express version starting at 2013 and maybe 2012
00:53:05sam686one way you could do is maybe rewind your virtual machine's clocks (not the host's clock) so it sees it as not yet expired
00:58:17raptorok, got a new license
00:58:22raptorhad to set up a microsoft account...
01:02:54sam686Try to rotate this level http://sam6.25u.com/upload/text1410/141017_00-02-37.txt
01:03:29sam686the level appears to disappears when rotates, because of the "Z" barrier
01:04:28sam686press Z and it zooms out to.. infinity
01:08:34raptorwell that's a bug...
01:08:40raptorgood find
01:08:53raptori thought i had fixed the rotate problems...
01:09:44sam686If you do criss cross goalzone it won't look good, and probable causes the centroid to go infinity/bad number
01:10:04sam686but on barriers it looks good with "Z" shape, it just don't rotate right.
01:10:20raptorok, can you please add that to the bug thread in the forums?
01:14:08sam686done in forums
01:14:12raptorthansk
01:17:14sam686im probably going to bed now, night
01:21:01raptornight!
01:23:14raptorphoew, ok solved that one..
01:23:25BFLogBot Commit: bd796d4ec6 | Author: buckyballreaction | Message: Don't throw a c++ exception in a Lua frame, instead use standard Lua error-raising mechanism. This let's Lua error messages be shown properly since moving to LuaJIT
01:26:46starseeker Quit (Ping timeout: 258 seconds)
01:30:53starseeker has joined
02:20:43raptorand good night!
02:20:45BFLogBot Commit: 0a2fd11172 | Author: buckyballreaction | Message: Minor speed improvements and notes for our findCentroid method
02:20:47raptor Quit ()
02:45:15Watusimoto has joined
03:15:56vs73nx has joined
03:15:56vs73nx Quit (Changing host)
03:15:56vs73nx has joined
03:53:40LordDVG has joined
04:20:12Watusimoto Quit (Ping timeout: 272 seconds)
04:41:53Darrel Quit (Ping timeout: 240 seconds)
04:42:46Darrel has joined
06:52:09vs73nx Quit (Quit: Quitting for now… See you all later)
07:16:30LordDVG Quit (Remote host closed the connection)
12:47:46Invisible has joined
13:13:58raptor has joined
13:13:58ChanServ sets mode +o
13:14:14raptorgood day!
13:16:43watusimoto has joined
13:16:43ChanServ sets mode +o
13:29:55Invisible Quit (Ping timeout: 272 seconds)
14:36:04Watusimoto_ has joined
14:41:58Watusimoto_ Quit (Ping timeout: 255 seconds)
14:51:33LordDVG has joined
15:14:36LordDVG Quit (Ping timeout: 246 seconds)
15:14:46Invisible has joined
15:18:57LordDVG has joined
16:13:53watusimotohi raptor!
16:14:02raptorhello!
16:14:07watusimotoI went to check in my edits last night and couldn't figure out which head to use
16:14:17watusimotoone was labeled "019 fixes"
16:14:22raptoryes, not that one
16:14:33raptori mean, that's the one sam686 and I are workign on
16:14:45watusimotook; it looked like there were fixes in both branches that would reasonably be part of the 019d release
16:14:55watusimotoso 019 fixes == 019d?
16:14:55raptoryes, I was transplanting back and forth
16:14:59raptorcorrect
16:15:01watusimotook
16:15:02watusimotogood
16:15:19raptoryour head should be 5761d75abd91
16:15:24watusimotoI'm not sure how I did it, but I seem to have resolved the issue of how to handle WallITems
16:15:29watusimotowe had that discussion last week or so
16:15:39raptoroh yeah?
16:15:40watusimotoand somehow the issue has just disappeared
16:16:00watusimotothere's still some special handling for walls, but less
16:16:35watusimotoI don't want to test too much, but everything seems to be working... client/server/editor
16:16:56watusimotoI may be ready for amgine to dig in this weekend if he still wants to do testing
16:17:11watusimotopressure test the editor
16:17:20raptorha
16:17:33watusimotothen give me reports like "it crashes"
16:17:38watusimoto"undo doesn't work"
16:17:42raptorok, you can push whenever
16:17:44watusimoto"something is broken"
16:17:47watusimototonight
16:17:52raptorok
16:18:31watusimotothe new undo design cleans some clutter out of uieditor
16:18:39raptoroh good
16:18:46raptorless clutter in there, the better
16:18:52watusimotoamen, brother
16:18:56raptorthat's gotta be our largest compilation unit...
16:19:04watusimotoover 5K lines
16:19:50watusimotough... still need to test undo/walls with plugins
16:21:26raptoryep, that's our largest - our 2nd largest is ship.cpp with 2825 lines
16:22:00watusimotoWe could pull out the editor menu ui which is in the same compilation unit
16:22:24watusimotothat would get rid of a few hundered rarely changing lines
16:22:31raptorhaha
16:22:44watusimotoI'e actually been meaning to do that for a while
16:24:03raptorthere are several code cleaning tasks I can think of - like removing rendering code from game objects
16:24:18raptorand consitent file CaseNaming
16:24:31raptorand slaying of random globals/externs
16:31:25sam686 Quit (Read error: Connection reset by peer)
16:34:33sam686 has joined
16:34:33ChanServ sets mode +v
16:41:30LordDVG Quit (Remote host closed the connection)
16:59:58raptorhere's a self-contradictory statement I just ran into: rapid development of web services using Maven and JAX-WS
17:15:37vs73nx has joined
17:16:46watusimotoprobably written by an MBA
17:17:44watusimotocompletely removing rendering code from objects could be difficult -- most details ahve alreayd been moved out to GameObjectRenderer
17:17:54watusimotobut the hooks are still in the object side
17:18:00raptorwhat i mean is the render() methods
17:18:14raptorwe'd have somthing akin to Joystick and JoystickRender
17:18:20watusimotoare you thinking a static render method -- render(wall); render(testItem); ...
17:18:35watusimotorather than testItem.render()?
17:18:49watusimotothat's probably the only way
17:19:07raptormostly I'm thinking about having to constantly fix dedicated builds because of missing #ifndef ZAP_DEDICATED in render methods on server classes
17:19:15watusimotoor create a set of renderer objects and assign them on the client side
17:19:17watusimotoi.e.
17:19:26raptoror even to get rid of ZAP_DEDICATED
17:19:29watusimototestItem.assignRenderer(testItemRenderer);
17:19:35raptorand then have a different main_server.cpp
17:19:53raptoryeah, i haven't thought too hard about it...
17:19:58watusimototestItem.render() --> calls testItem.renderer->render()
17:20:10watusimotobut on server, testItem.renderer is null
17:20:38raptorsince the render methods ar on the client, i was thinking of just doing: TestItemRender->render()
17:20:39watusimotozap_dediated must die
17:20:56raptorand it would have access internal TestItem object
17:21:01watusimotoTestItemRenderer->render(obj)
17:21:04watusimotoyou mean
17:21:22raptorhmmm... i'm thinking along static object lines, i think
17:21:26raptorwhich won't work here
17:21:33raptoryeah, so i haven't thought much about it yet... :)
17:21:36watusimotothat last one I wrote would be static
17:22:06watusimotothough you don't want to have to think about which function renders which object type... or else you get a huge case statement
17:22:28raptorcorrect
17:22:31watusimotomaybe the best way is to have objects still have a .render() method
17:22:50raptorbut then we need to keep ZAP_DEDICATED
17:23:05watusimotono... like this:
17:24:01watusimotoclass TestItem() has a pointer to a render function
17:24:06watusimotoon client:
17:24:29watusimotoTestItem::setRenderFunction(TestItemRenderer)
17:24:32watusimotoon server:
17:24:48watusimotoTestITem::setRenderFunction(null)
17:24:56watusimotoor better, on server
17:25:18watusimotoTestItem::setRenderFunction({TNLAssert(false, "No rendering on server!")})
17:25:26watusimoto(if you could use lambdas in c++)
17:25:41raptorc++11 allows lamndas
17:25:50raptorbut yes, i see
17:25:51watusimotothen you can just do TestItem.render()
17:25:57watusimotoand you get an assert on server
17:26:19watusimotobut the call itself would be valid anywhere
17:26:29watusimotojust so long as you only use it on the client
17:27:09watusimotothen you can get rid of some ifdefs, I think
17:27:38raptorhmmm
17:39:26watusimotowe could try that on a single object, to see how it works
18:11:36vs73nx Quit (Remote host closed the connection)
18:12:08vs73nx has joined
18:12:08vs73nx Quit (Changing host)
18:12:08vs73nx has joined
18:17:27Invisible Quit (Ping timeout: 272 seconds)
18:27:50Watusimoto_ has joined
19:03:20raptorheading home, later!
19:03:31raptor Quit ()
20:18:05vs73nx Quit (Quit: Quitting for now… See you all later)
21:01:12fordcars has joined
21:11:54Watusimoto_ Quit (Remote host closed the connection)
22:06:34watusimoto Quit (Quit: Leaving.)
22:20:40watusimoto has joined
22:20:40ChanServ sets mode +o
22:50:28watusimoto Quit (Quit: Leaving.)
22:54:07koda has left #bitfighter
23:58:13sam686raptor last visited forum: 3 minutes ago

Index Search ←Prev date Next date→

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