Timestamps are in GMT/BST.
| 00:22:35 | | Watusimoto has joined |
| 00:33:53 | Watusimoto | bf.org still down :-( |
| 00:34:21 | fordcars | Yep :( |
| 00:51:38 | raptor | no ping yet.. |
| 01:27:39 | | raptor Quit () |
| 01:37:10 | | raptor has joined |
| 01:37:10 | | ChanServ sets mode +o |
| 01:43:12 | | Watusimoto_ has joined |
| 01:45:25 | | YoshiSmb_m has joined |
| 01:46:01 | | Watusimoto Quit (Ping timeout: 248 seconds) |
| 01:50:46 | | Watusimoto_ Quit (Ping timeout: 240 seconds) |
| 01:55:11 | YoshiSmb_m | Speedy Gonzalez, the fastest mouse in Mexico |
| 01:55:43 | YoshiSmb_m | *Gonzales |
| 01:58:25 | raptor | I think I remember watching that when I was really young |
| 01:59:10 | YoshiSmb_m | yea, good times |
| 01:59:39 | YoshiSmb_m | i just watched lonely toons twice a day |
| 02:01:40 | YoshiSmb_m | Act 1: Micky Mouse and Speedy Gonzales started to fight |
| 02:02:03 | YoshiSmb_m | Act 2: Jerry Joined the Fight |
| 02:03:26 | YoshiSmb_m | Act 3: a Normal Mouse Joined the fight too |
| 02:03:57 | YoshiSmb_m | How it's the act called? |
| 02:04:16 | raptor | i don't understand |
| 02:06:21 | YoshiSmb_m | 1: Mickey Mouse started to fight against Speedy Gonzales |
| 02:06:44 | YoshiSmb_m | 1:Jerry joined the fight |
| 02:06:48 | YoshiSmb_m | *2: |
| 02:08:08 | YoshiSmb_m | 3: A normal mouse joined the fight too |
| 02:08:30 | YoshiSmb_m | How it's called the act? |
| 02:09:09 | raptor | oh, i didn't understand the question - do you mean: why is it called an 'act' ? |
| 02:11:09 | YoshiSmb_m | nevermind |
| 02:26:41 | | YoshiSmb_m Quit (Remote host closed the connection) |
| 02:30:31 | | amgine123 has joined |
| 02:30:44 | amgine123 | hi raptor |
| 02:30:51 | raptor | hello |
| 02:31:18 | amgine123 | so now that i have my BF clone setup anything a begginier can do to help ;) |
| 02:32:38 | raptor | did you get it to build? |
| 02:34:38 | amgine123 | yes |
| 02:35:43 | raptor | oh cool |
| 02:35:56 | raptor | good job |
| 02:36:05 | raptor | what IDE did you use? Visual Studio? |
| 02:37:29 | amgine123 | codeblocks v |
| 02:37:44 | amgine123 | codeblocks IDe |
| 02:38:49 | raptor | ah ok |
| 02:38:57 | raptor | i don't think i've ever used that one on Windows |
| 02:42:10 | | amgine123 Quit (Ping timeout: 246 seconds) |
| 02:51:45 | | raptor Quit () |
| 03:32:03 | fordcars | It hurts my eyes: http://mindprod.com/jgloss/unmainnaming.html |
| 03:35:29 | | CrazyLinuxNerd has joined |
| 04:33:13 | | raptor has joined |
| 04:33:13 | | ChanServ sets mode +o |
| 04:33:19 | raptor | fordcars: that's hilarious! |
| 04:33:47 | fordcars | I'm still reading it haha |
| 04:34:19 | raptor | very quotable: "Real men never define acronyms" |
| 04:34:23 | raptor | haha |
| 04:34:29 | fordcars | Hahahahahaha |
| 04:34:46 | fordcars | Yeah, I laughed so much in this :,) |
| 04:37:31 | raptor | i'm starting at the beginning of the essay |
| 04:37:42 | raptor | it's funny because... i'm usually a maintenance programmer |
| 04:38:19 | fordcars | Haha, the essay is literally made to make maintenance programmers cry |
| 04:38:35 | fordcars | ... or go through the whole code |
| 04:40:56 | raptor | hahahaha |
| 04:41:15 | raptor | this is so funny... because I'm even a java maintainer |
| 04:41:22 | raptor | i can't stop laughing.. |
| 04:42:43 | raptor | "no academic would be caught dead giving an example" |
| 04:44:47 | fordcars | Hahahah |
| 04:45:39 | fordcars | I think I'll keep these quotes and hang them in my room |
| 04:45:55 | raptor | "Choose variable names that masquerade as mathematical operators" |
| 04:46:03 | raptor | openParen = ( slash + asterix ) / equals; |
| 04:46:10 | fordcars | That is awful hehe |
| 04:53:27 | fordcars | a=a++; or f(a++,a++); |
| 04:53:40 | raptor | oh man |
| 04:53:58 | fordcars | Haha, that is so weird |
| 06:14:49 | | YoshiSmb_m has joined |
| 07:42:30 | | YoshiSmb_m Quit (Remote host closed the connection) |
| 08:02:05 | | raptor Quit () |
| 08:55:15 | | Nothing_Much Quit (Read error: Connection reset by peer) |
| 08:57:29 | | Nothing_Much has joined |
| 10:20:37 | | Akien has joined |
| 10:20:47 | Akien | Hi there |
| 10:27:40 | | Watusimoto has joined |
| 10:44:30 | fordcars | Hi |
| 10:50:23 | Akien | As pointed out in https://code.google.com/p/bitfighter/issues/detail?id=444 (though the issue is about something else for the master branch), I'm having troubles packaging bitfighter 019c |
| 10:51:04 | Akien | For some reason the installed version looks for music and scripts in ~/.bitfighter/{music,scripts} instead of /usr/share/games/bitfighter/{music,scripts} |
| 10:51:34 | Akien | It does seem to find the rest of the data in /usr/share/games/bitfighter, since I can play the game with graphics and so on |
| 10:53:26 | Akien | (I meant packaging for Linux of course :) ) |
| 11:05:44 | Akien | Ah well, I just rebuilt my 019c package (I had successfully packaged a snapshot of 020 in the meantime), and now it works |
| 11:06:21 | Akien | It seems the game copies music and scripts from the LINUX_DATA_DIR to ~/.bitfighter at the first launch |
| 11:11:59 | | fordcars Quit (Quit: Page closed) |
| 11:14:54 | Akien | Ok so yes, looking at the code there are some "first launch" routines that are used only if ~/.bitfighter does not exist yet |
| 11:17:11 | Akien | So I don't know what made that I had a ~/.bitfighter without the music and resources, but I guess it wouldn't hurt to retry copying the resources when they can't be located in an existing ~/.bitfighter folder |
| 11:17:18 | Akien | I'll file an issue about it :-) |
| 11:23:12 | | Watusimoto Quit (Ping timeout: 264 seconds) |
| 11:49:38 | | Watusimoto has joined |
| 12:05:01 | | Watusimoto Quit (Ping timeout: 252 seconds) |
| 12:19:21 | | Watusimoto has joined |
| 13:06:27 | | Watusimoto Quit (Ping timeout: 245 seconds) |
| 13:46:13 | | Watusimoto has joined |
| 15:10:30 | | Watusimoto Quit (Ping timeout: 255 seconds) |
| 17:50:19 | | Watusimoto has joined |
| 17:57:44 | Akien | Bitfighter is now available in Mageia: http://svnweb.mageia.org/packages/cauldron/bitfighter/current/SPECS/bitfighter.spec?view=markup :-) |
| 18:07:03 | | raptor has joined |
| 18:07:03 | | ChanServ sets mode +o |
| 18:07:12 | raptor | good morning! |
| 18:07:55 | raptor | Akien: thanks for packaging! |
| 18:08:15 | Akien | raptor: Morning :-) |
| 18:08:16 | raptor | yes, we have some first-launch routines that copy many of the game resources to the ~/.bitfighter directory |
| 18:08:46 | raptor | we made a choice to do music/scripts/plugins there, too, because people like to add their own music |
| 18:09:20 | raptor | perhaps one thing we should consider is to do some sort of virtual filesystem that aggregates game data from 2 different places |
| 18:09:37 | raptor | like physfs |
| 18:10:17 | raptor | i didn't know mageia was still in development |
| 18:11:20 | Akien | raptor: Yes Mageia is alive and kicking, with a quite strong community nowadays |
| 18:11:27 | raptor | oh good |
| 18:11:41 | Akien | We released Mageia 4 in February, Mageia 5 should be there (with bitfighter!) in December |
| 18:11:57 | raptor | oh? we made it into a standard release!? |
| 18:12:17 | Akien | Yes I'm an offical Mageia packager :) |
| 18:12:23 | raptor | that's something we've been hoping for one day... :) |
| 18:13:50 | Akien | raptor: For the music maybe you could load the content of ${datadir}/music first and then the user content from ~/.bitfighter/music? |
| 18:13:57 | raptor | yes |
| 18:14:00 | raptor | that is a good idea |
| 18:14:19 | raptor | and do the same for scripts/etc. |
| 18:14:26 | Akien | Yep |
| 18:16:08 | raptor | looking at your new issue... |
| 18:22:11 | Akien | I'm getting a crash when changing the display from "Fullscreen" back to "Windowed" in the menu UI. Not exactly a crash, but the game disappears and seems to be in zombie mode (I have to kill it manually) |
| 18:22:43 | Akien | The crash is not present with the master branch, so I'm looking at the commit logs for a potential fix :-) |
| 18:22:55 | raptor | opened issue https://code.google.com/p/bitfighter/issues/detail?id=446 |
| 18:23:08 | raptor | Akien: did you compile against SDL 1.2 or SDL2 ? |
| 18:23:24 | Akien | raptor: Nice thanks! :-) |
| 18:23:30 | Akien | Against SDL2 |
| 18:23:40 | raptor | hmmm... |
| 18:24:04 | raptor | the only thing we changed in master is to default to SDL2 for Linux; before was SDL 1.2 (I think) |
| 18:24:13 | Akien | With the master branch it seems the whole UI is redrawn when switching from Fullscreen to Windowed mode |
| 18:24:34 | Akien | But somehow this step fails with 019c |
| 18:25:30 | raptor | let me check your SPEC file again.. |
| 18:26:12 | raptor | looks OK to me |
| 18:26:38 | Akien | To be sure I tried a build without my patches to use system alure and boost |
| 18:27:26 | Akien | But the same thing happens |
| 18:27:41 | raptor | yeah, we always link against system SDL (no in-tree copy) |
| 18:28:23 | Akien | Ah wait a second |
| 18:28:44 | Akien | I think I actually built it against SDL1.2 |
| 18:28:45 | raptor | i'm not finding anything since 019c that we've done to our UI system with regards to SDL |
| 18:28:58 | raptor | oh, run: ldd bitfighter |
| 18:29:06 | raptor | err, on the executable |
| 18:29:45 | Akien | Yep that's it, the spec file I linked is the one I pushed on Mageia's buildsystem, but actually I had forgotten to set USE_SDL2=ON at first |
| 18:29:54 | Akien | So my local verison was built against SDL1.2 |
| 18:31:53 | raptor | ok, yeah - there's a known SDL 1.2 bug that freezes the window and spikes the CPU on some window systems |
| 18:32:38 | raptor | but only on a window change |
| 18:32:54 | | kaen has joined |
| 18:33:44 | Akien | raptor: So yes it works fine with SDL2 |
| 18:33:57 | raptor | oh good, excellent |
| 18:34:28 | raptor | we've already moved to SDL2 for Windows/OSX, but I still maintain SDL 1.2 for *nix |
| 18:36:38 | Akien | It's a good thing, I guess some distros haven't packaged SDL2 yet (or they have, but they still support previous version without SDL2 in the repos) |
| 18:38:54 | kaen | cool, you even packaged the notifier :) |
| 18:39:03 | kaen | I didn't even remember to do that for my debian package :P |
| 18:39:12 | kaen | nice woek |
| 18:39:14 | kaen | work |
| 18:40:54 | Akien | I'll admit I took the idea from OpenSUSE's package :-) https://build.opensuse.org/package/view_file/games/bitfighter/bitfighter.spec |
| 18:41:41 | raptor | good morning kaen |
| 18:41:49 | kaen | ah, I see. raptor, is that suse package yours? |
| 18:41:49 | raptor | i package for openSUSE :) |
| 18:41:51 | kaen | morning all |
| 18:42:02 | raptor | yep |
| 18:42:08 | kaen | guess I should have done the same :P |
| 18:42:35 | kaen | I ordered parts for a new rig today :) |
| 18:42:44 | kaen | "superb" tier at http://www.logicalincrements.com/ |
| 18:42:45 | raptor | the openSUSE package was always privileged - it had a developer do it... |
| 18:43:00 | Akien | raptor: You might be interested in my patch for the notifier then, because as far as I can tell, it would lack the icon in OpenSUSE |
| 18:43:02 | kaen | much better than my 1.8ghz dual core craptop |
| 18:43:17 | raptor | Akien: i'll take a look.. |
| 18:43:24 | raptor | a new compy! |
| 18:43:35 | kaen | I get the same problem on debian if it's not invoked from the right directory |
| 18:43:58 | raptor | ah yes - yep, broke |
| 18:44:03 | Akien | Yep, that fixes it: http://svnweb.mageia.org/packages/cauldron/bitfighter/current/SOURCES/bitfighter-019c-mga-notifier-icon.patch?revision=640648&view=markup |
| 18:44:03 | raptor | that's my fault for sure |
| 18:44:07 | Akien | It's a hack though |
| 18:44:28 | Akien | (that fixes only for Mageia actually, because I put the icon in /usr/share/icons/hicolor/... |
| 18:45:05 | raptor | maybe that's the correct fix though - it is just a python script |
| 18:45:30 | raptor | maybe I should provide a variable at the top that points to the icon directory |
| 18:45:37 | raptor | or maybe... embed the the icon! |
| 18:46:01 | Akien | Ideally their should be "make install" instructions, and the icon path would be retrieved from the one set for the install |
| 18:46:21 | Akien | s/their/there/ |
| 18:46:35 | raptor | hmm.. so add it to the cmake system |
| 18:46:43 | Akien | Yes |
| 18:47:25 | Akien | So that it always uses the main icon, be it in /usr/share/pixmaps or /usr/share/icons/hicolor or whatever standard path :-) |
| 18:48:05 | raptor | I think I avoided 'make install' for Linux because of all the various distros wanting things in different places |
| 18:48:17 | raptor | I don't fully remember, though |
| 18:48:43 | raptor | be right back.. |
| 18:48:46 | Akien | Yes that's true, but I think you should provide a default "make install" to put things in /usr/local, and cmake arguments to make it easy to change that |
| 18:51:34 | Akien | Basically define paths such as BF_BINARY_INSTALL_DIR = ${CMAKE_INSTALL_PREFIX}/bin and BF_DATA_INSTALL_DIR = ${CMAKE_INSTALL_PREFIX}/share/ |
| 18:52:24 | Akien | And make it sure that they can be overridden with "cmake -DBF_BINARY_INSTALL_DIR=/usr/games -DBF_DATA_INSTALL_DIR=/usr/share/games" for example |
| 18:55:27 | Akien | Then for the icon, desktop file and man page you could directly use e.g. ${CMAKE_INSTALL_PREFIX}/share/man/man1/ |
| 18:56:19 | | Watusimoto Quit (Ping timeout: 240 seconds) |
| 19:02:09 | kaen | yep, we could slap into cmake and it would install sanely on all distros and even windows |
| 19:04:33 | raptor | preposterous! |
| 19:10:14 | raptor | and another: https://code.google.com/p/bitfighter/issues/detail?id=447 |
| 19:10:41 | Akien | Nice :-) |
| 19:11:36 | raptor | kaen: i don't know if you look at the logs; but, I found the real cause to our bitfighter_client problems |
| 19:11:44 | raptor | with the build system |
| 19:12:11 | raptor | if you're interested, my monologue starts here: http://bitfighter.org/irclogs/index.php?date=2014-06-27#l127 |
| 19:15:17 | | LordDVG has joined |
| 19:17:35 | kaen | looks good to me |
| 19:17:43 | kaen | I didn't know cmake had such a thing |
| 19:17:55 | kaen | but that is definitely exactly what we need |
| 19:20:54 | raptor | I also had this idea of making the _shared objects their own library so we don't ahve to recompile that between dedicated/normal server |
| 19:21:03 | raptor | but then I found out we compile with ZAP_DEDICATED |
| 19:21:34 | kaen | ah, right |
| 19:21:43 | kaen | dedicated build is cheap anyway |
| 19:21:46 | raptor | and so many objects have their render() methods defined out by that.. |
| 19:21:49 | kaen | even on my laptop, I consider it "fast" |
| 19:21:53 | raptor | haha |
| 19:22:10 | kaen | man I'm so pumped for my new rig |
| 19:22:13 | raptor | i wonder if we should pull out the render stuff into things like LineItemRender |
| 19:22:25 | kaen | I've always wondered what it feels like to run make -j8 |
| 19:22:30 | raptor | hahaha |
| 19:22:34 | raptor | it feels glorious! |
| 19:22:41 | raptor | my work computer can do that |
| 19:24:05 | kaen | I've thought about pulling out the render stuff before, too |
| 19:24:20 | kaen | I'm not convinced we'd get a good return on the effort though |
| 19:24:30 | raptor | yeah, that's true |
| 19:30:16 | Akien | Do you know if SDL2 can support unusual key mappings such as &, é or " ? |
| 19:30:22 | Akien | The French keyboard is somewhat funny: http://fr.wikipedia.org/wiki/AZERTY#mediaviewer/Fichier:Clavier-Azerty.svg |
| 19:30:41 | Akien | And the numbers can't be access directly, you have to press Shift + 1 to get "1", if not you get & |
| 19:30:56 | kaen | whoa, that is pretty funny |
| 19:31:06 | Akien | So the 1, 2, 3 key mapping to change weapons in bitfighter don't work, and can't be redefined to the corresponding &, é, " keys |
| 19:31:25 | kaen | our own key handling is pretty weird. we have our own constants (taken from SDL constants?) that we use throughout the codebase |
| 19:31:33 | Akien | Yes that's a pain for most games, you always have to redefine WASQ and numbers :-) |
| 19:32:11 | kaen | do you have such a keyboard? what happens if you try to map something to é? |
| 19:32:31 | Akien | Yes my keyboard is like that |
| 19:32:42 | Akien | With "é" I get an empty box |
| 19:32:53 | Akien | For & and " I get "unknown key" |
| 19:33:05 | kaen | I think that's our fault, not sdl's |
| 19:33:33 | Akien | kaen: Well yes I just tried and the key mapping works ingame |
| 19:33:39 | Akien | s/ingame/in-game/ |
| 19:34:03 | kaen | interesting. it just doesn't show the character é? |
| 19:34:23 | Akien | So it would probably be a matter of adding new SDL constants for exotic keyboard |
| 19:34:41 | kaen | actually, I think I can change my input method and play around with it |
| 19:34:44 | Akien | The three keys work actually |
| 19:35:04 | Akien | Yes with "setxkbmap fr" most likely |
| 19:38:58 | kaen | yep, works well enough |
| 19:39:09 | kaen | I guess the problem is in the keycode -> character mapping |
| 19:40:15 | kaen | iirc there's a unicode flag we can set for sdl to make it grok those characters |
| 20:07:27 | Akien | Ah nice, it seems bitfighter is the 100th package that I maintain for Mageia :-) http://people.mageia.org/u/akien.html |
| 20:07:37 | Akien | You won... a nice maintainer :-p |
| 20:09:53 | raptor | oh wow |
| 20:09:56 | raptor | that's a ton! |
| 20:10:17 | raptor | is Mageia a predominantly French distribution? |
| 20:14:59 | Akien | It's an international community, but it's true that a good share of the contributors are French (a quarter or a third I'd say, it depends on how much translators we have) |
| 20:15:02 | raptor | kaen: was it intentional to not have a CMakeLists.txt for poly2tri and compile it as a separate lib? |
| 20:16:13 | raptor | i noticed that clipper has it's own, but not poly2tri |
| 20:16:22 | kaen | yes, but only because it's not packaged in any distros yet |
| 20:16:55 | Akien | It is in Mageia :-) But I packaged it two weeks ago or so, it's true that it's quite new :-) |
| 20:17:08 | kaen | oh, cool :) |
| 20:17:28 | Akien | I would be pleased if there was a possibility to use the system poly2tri in the next release |
| 20:17:37 | raptor | ok, i'll add it |
| 20:17:50 | raptor | i'm working on migrating to the OBJECT libraries |
| 20:17:51 | Akien | BTW I wondered, it's probably a matter of taste but wouldn't it make sense to put all external libraries in a subfolder? |
| 20:18:00 | Akien | external/ or thirdparty/ or something? |
| 20:18:04 | raptor | oh yes |
| 20:18:14 | raptor | it would make a lot of sense - our root dir is a mess |
| 20:18:24 | Akien | :-D |
| 20:35:19 | raptor | oh lovely, the windows ZAP.rc file can only be added onto the final exe, and not into the client lib |
| 20:38:35 | raptor | oh no... all forcefields point to the right now.. |
| 20:39:12 | raptor | and walls are missing! |
| 20:39:20 | raptor | and turrets point to the right! |
| 20:42:13 | raptor | i guess just polywalls |
| 20:42:48 | Akien | :-D |
| 20:43:19 | raptor | yeah, i recommend not using the code from master for a little while... |
| 20:50:13 | raptor | i have to determine if it was me or watusimoto that broke it... |
| 20:52:44 | sam686 | kaen, notice that your buildbot died again? |
| 20:53:10 | raptor | and the winner is... |
| 20:53:17 | raptor | watusimoto! |
| 20:56:32 | sam686 | https://twitter.com/URPad it just keep getting some updates about 2 times a day |
| 20:57:07 | sam686 | They say most is online, but looks like not bitfighter.. |
| 20:58:51 | raptor | i hope wer're not on LABL5,6,7 |
| 20:58:56 | raptor | *we're |
| 21:00:17 | sam686 | labl7 probably is a complete loss of no backup |
| 21:00:27 | raptor | yeah... eeek |
| 21:01:02 | raptor | sam686: can you tell me again why you made that recent ALURE change? |
| 21:01:13 | raptor | i'm thinking it may be a build system issue |
| 21:01:43 | raptor | if so, my next commit should fix it - otherwise we may want to add that fix to upstream |
| 21:02:03 | sam686 | it crash on quitting bitfighter, empty list causes vector .begin() to crash it appears, don't know why so I put an empty list check |
| 21:02:18 | raptor | ah ok, when exiting... let me test that |
| 21:02:30 | sam686 | on windows it does crash |
| 21:02:39 | sam686 | linux might be different maybe... |
| 21:03:38 | raptor | yeah, i think it was a build system memory alignment problem |
| 21:27:49 | kaen | I like urpad's twitter backdrop. "reliable hardware" is the biggest bubble |
| 21:27:55 | kaen | fake it 'til you make it |
| 21:28:09 | raptor | hahaha |
| 21:28:29 | raptor | did you read what actually happened to them? |
| 21:29:21 | kaen | I did not. where can I read tha? |
| 21:29:23 | kaen | that? |
| 21:29:26 | raptor | the out-of-band server management software on the blades (IPMI implementation) was hacked |
| 21:29:32 | kaen | ooo |
| 21:29:44 | raptor | let me find it... |
| 21:30:25 | | Watusimoto has joined |
| 21:30:49 | kaen | that explains the lack of a "reliable software" bubble |
| 21:31:17 | raptor | kaen: forwarded you the e-mail Watusimoto had forwarded to me... i'm not sure if the info was made public |
| 21:31:31 | kaen | ok, thanks |
| 21:32:31 | raptor | i committed my cmake OBJECT lib changes |
| 21:32:43 | raptor | may older distro maintainers forgive me... |
| 21:33:29 | sam686 | This code will crash horribly with some memory corruption, on windows, expecially when placed in bitfighter main(). http://sam6.25u.com/upload/text1406/140628_12-32-23.txt |
| 21:33:59 | raptor | sam686: sounds like you need to 'make clean' :) |
| 21:34:08 | sam686 | it might be possible that the order of destructor during program exit is what might crash it. |
| 21:34:38 | sam686 | its nothing to do with make clean stuff, its accessing list begin() after delete is whats causing the problem |
| 21:34:44 | raptor | i just pushed my changes, please try a full clean build with them, including redo-ing cmake |
| 21:34:57 | kaen | that's definitely the problem :P |
| 21:35:09 | kaen | you're dereferencing a delete'd pointer and then invoking a method on it |
| 21:35:14 | raptor | yes, i think it was a memory alignment problem with linking static libs in our build system |
| 21:35:39 | kaen | I must be missing some context. what does that show, sam686 ? |
| 21:35:52 | raptor | yeah that is sam686's problem in his example; but he is referring to his ALURE fix which I just reverted, which I think was due to our build system |
| 21:36:16 | kaen | ah, I see |
| 21:37:06 | raptor | with my just-committed changes, I did a full recompile and there was no crash (windows xp with vc++ 2010) |
| 21:37:13 | sam686 | Even doing zzz->~list(); instead of delete, still causes crash on accessing begin() afterwards |
| 21:38:55 | raptor | yes that is definitely the problem |
| 21:38:59 | raptor | with your example |
| 21:39:06 | raptor | but i don't think there is a problem with ALURE |
| 21:39:29 | raptor | i think it was a side-effect of our weird linking in our build system on windows |
| 21:39:33 | raptor | which I hope i just fixed |
| 21:42:05 | kaen | man, I haven't written c++ for ages it feels like. my brain is becoming tragically ruby-fied |
| 21:43:09 | kaen | I've actually been writing tons of bash for work |
| 21:43:17 | raptor | fordcars recently showed me this: http://mindprod.com/jgloss/unmain.html |
| 21:43:41 | raptor | oh yeah... bash would be king at a devops type job |
| 21:45:25 | kaen | I hope it makes a comeback. The big revolution in devops right now is Docker, which is overtaking the old king Chef |
| 21:45:34 | kaen | which translates to a shift from ruby to bash |
| 21:45:42 | raptor | i've never even heard of that |
| 21:47:04 | kaen | docker has been out for just over a year. chef I think is two or three years old. I guess I might have tunnel vision, but that's the case with the startups I work with |
| 21:47:22 | raptor | and i've never even heard of chef either... |
| 21:47:32 | raptor | except in some of the github repos of that company that hired you |
| 21:48:03 | kaen | they're pretty cool systems. basically you write up "recipes" to build a web, db, jobs, solr, etc. server |
| 21:48:19 | kaen | then you can deploy those to arbitrary servers elsewhere |
| 21:48:31 | kaen | so you like codify your infrastructure for 100% automation |
| 21:48:47 | kaen | my day-to-day is spinning up EC2 instances and deploy chef recipes to them |
| 21:48:55 | kaen | and now, converting those recipes to docker |
| 21:49:02 | raptor | so is it some abstract system that uses the native system of the OS to implement its commands? |
| 21:49:08 | kaen | chef is, yes |
| 21:49:24 | kaen | it's abstracted so you can deploy to ubunutu, rhel, or even a windows server |
| 21:49:54 | kaen | docker, however, is like bare metal. it invokes steps through sh |
| 21:49:55 | sam686 | on release build, it still crashes on alure stopstream when I try to enter editor, but exits normally without crash when selecting "exit" or closing the window |
| 21:50:22 | sam686 | on debug build, a bunch of other errors trying to enter editor |
| 21:50:26 | kaen | so it's not as general-purpose, but you just right bash (or whatever) instead of learning a new chef-specific DSL |
| 21:50:32 | kaen | just write* |
| 21:50:39 | raptor | sam686: watusimoto is in the middle of breaking lots of things |
| 21:50:53 | raptor | it's why polywalls don't work at the moment... |
| 21:51:16 | sam686 | http://sam6.25u.com/upload/text1406/140628_12-51-08.txt |
| 21:51:20 | kaen | whoa, that's a deep level of breakage |
| 21:51:49 | raptor | ah, sam686, see the backtrace part with Zap::Level:: |
| 21:52:01 | raptor | that's Watusimoto's new code |
| 21:52:21 | raptor | so i suspect that whatever he did caused a memory bug that manifested itslef as that ALURE problem |
| 21:52:57 | raptor | I've decided to just stay out of the way, at the moment, and work on build issues or minor compiler issues |
| 21:53:02 | kaen | I think I get it |
| 21:53:16 | kaen | something in level caused a normal-ish error, then in backtrace handler gets invoked |
| 21:53:31 | raptor | ah.. yeah |
| 21:53:34 | raptor | that makes sense |
| 21:53:37 | kaen | and alure is in some state where it didn't expect the abort/exit sequence, and segfaults or whatever |
| 21:53:48 | sam686 | This is when exiting normally http://sam6.25u.com/upload/text1406/140628_12-53-36.txt |
| 21:53:52 | raptor | yes, that makes total sense |
| 21:54:28 | sam686 | the main difference is there a "dynamic atexit destructor" which means we don't know what order it is destructing the4 globals |
| 21:55:12 | raptor | sam686, I think I'd just let Watusimoto finish his Level work first, then we should run valgrind or something and start polishing things up |
| 21:55:21 | kaen | I think once that abort handler is triggered, all bets are off from alure's point of view |
| 21:55:28 | kaen | I agree with raptor, too |
| 21:55:48 | raptor | right now i don't really trust the errors much because I think there's a lot of fun memory problems |
| 21:56:13 | raptor | back in a few min.. |
| 21:56:50 | sam686 | maybe thats not the reason... http://sam6.25u.com/upload/text1406/140628_12-56-26.txt without SoundSystem::shutdown(); |
| 21:57:16 | sam686 | Probable just corruption from bad way of entering level editor |
| 21:57:47 | sam686 | it only crash when entering level editor... |
| 22:11:00 | raptor | ok, i'll test in the editor |
| 22:11:24 | raptor | ha, there's an assert there |
| 22:11:52 | raptor | actually, i can't even enter the level editor |
| 22:12:06 | raptor | segfault after assert with Level object stuff |
| 22:12:19 | raptor | so i'll just wait for watusimoto's polishing |
| 22:22:44 | Watusimoto | editor is hopelessly broken. I fixed on problem earlier, will try to get it working soon |
| 22:28:05 | sam686 | There is problems with stack tracer... 1: it doesn't catch integer division by zero |
| 22:28:23 | sam686 | 1. its best to disable stack tracer in debug.. |
| 22:28:30 | sam686 | or that was number 2.. |
| 22:29:13 | sam686 | for example, int zz[2]; zz[0] = zz[888888888]; hopelessly terminates the program immediately without halting in debug build |
| 22:29:47 | sam686 | and the stack traser looks like a whole bunch of unreadable stuff http://sam6.25u.com/upload/text1406/140628_13-27-02.txt |
| 22:30:30 | sam686 | Not even the correct back trace either |
| 22:31:39 | sam686 | oh i see, theres main() at the bottom thats part is correct |
| 22:32:03 | sam686 | its just adding so much junk its hard to read |
| 22:34:29 | raptor | hi Watusimoto |
| 22:34:43 | raptor | I made some significant cmake changes - you'll want to rerun it |
| 22:38:19 | sam686 | I found that if exit is removed from abortHandler, it will just go back to where the error is and crash like if there is no stack tracer (after printing/displaying stack trace) |
| 22:41:25 | Watusimoto | you can write tests! |
| 22:41:58 | Watusimoto | (reading through the comments here as I wait for penalty kicks in the chile v brasil game |
| 22:42:22 | raptor | looks like barrier.h has gtest code in it |
| 22:42:28 | raptor | that should be removed |
| 22:42:51 | Watusimoto | hi raptor :-) |
| 22:43:41 | Watusimoto | this level refactor has been the biggest of my life, in or out of work |
| 22:43:49 | raptor | also stringUtils.cpp uses renderutils now? |
| 22:43:54 | raptor | ok |
| 22:44:13 | Watusimoto | I think renderutils has a fn that could be moved to string utils... I don't remember the details |
| 22:46:16 | Watusimoto | I'm wondering if we should refactor the way clientGame works -- now we are always in a client game, even when not playing, but only in a servergame when hosting. Maybe clientgame should only exist while playing, and uimanager should be the main thing running all the time |
| 22:46:38 | Watusimoto | this would make some of the weirdness I'm seeing in the editor go away |
| 22:46:43 | Watusimoto | just a thought for later |
| 22:47:54 | raptor | yeah, I think it's a good idea - it seems like clientgame has lost its original purpose (whatever it was) |
| 23:01:35 | kaen | clientgame is more like ClientApp, and GameType is more like what a Game class would be |
| 23:01:55 | kaen | that really messed with my understanding when I first joined |
| 23:03:27 | kaen | zap really had a broken class hierarchy that I think bitfighter has been hobbling on ever since |
| 23:05:48 | raptor | it probably broke because of the amateurish additions of myself before you joined... :) |
| 23:08:01 | kaen | no, I got curious one day and looked at an ancient zap version |
| 23:08:33 | raptor | oh? the hierarchy still had some of the current failings? |
| 23:08:41 | kaen | most of the problems started before bitfighter was a twinkle in watusimoto's eye |
| 23:11:04 | kaen | in fact, I remember seeing the same goofy clientgame/servergame mess in the tnl tutorial game |
| 23:14:49 | sam686 | int *z = new int[10]; for(int a=0; a < 10000000; a++) z[a] = 0; does get caught by stacktracer, but fails to fprintf due to another crash |
| 23:16:19 | kaen | I think sam's right about disabling stacktrace in debug mode |
| 23:18:30 | raptor | I need a good variable name to force in-tree library compiling |
| 23:18:44 | raptor | FORCE_IN_TREE_SQLITE seems too long |
| 23:19:15 | sam686 | printf doesn't crash, fprintf does crash when memory is corrupted causing any new malloc to fail |
| 23:19:36 | raptor | or maybe just: SQLITE_INTREE=1 |
| 23:19:42 | sam686 | and fprintf is calling malloc |
| 23:20:04 | kaen | FORCE_IN_TREE_SQLITE seems fine to me |
| 23:20:28 | kaen | doesn't really matter though, as long as we're consistent |
| 23:21:30 | sam686 | umm... no logbot commit message? |
| 23:21:40 | kaen | I just downloaded zap again to verify my memory, and indeed it comes complete with clientgame, servergame, gametype, and the class-to-filename mismatches |
| 23:22:13 | raptor | i remember that when you had to tell early command-line linux CD writers to do something weird like write at a different sector size, it would make you specify a command options like this: --use-the-force-luke=override-sector |
| 23:22:18 | sam686 | probably missing commit php file in bitfighter.org website |
| 23:23:31 | raptor | I thought I had committed that web hook somewhere... |
| 23:23:57 | kaen | lol |
| 23:24:03 | sam686 | error C1083: Cannot open include file: 'sqlite3.h': No such file or directory when building master |
| 23:24:06 | kaen | I saw the other day that mysql has a --i-am-a-dummy flag |
| 23:24:11 | raptor | I'm fixing master now! |
| 23:24:47 | sam686 | (4158a13e7b82) Disable StackTrace during debug builds (logbot fails) |
| 23:25:10 | raptor | that was a good idea sam686 |
| 23:25:38 | | BFLogBot some text raptor |
| 23:25:40 | | BFLogBot another commit |
| 23:25:41 | | BFLogBot try3 still try3 still try 3fdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafda |
| 23:25:43 | | BFLogBot sfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsaf |
| 23:25:44 | | BFLogBot dsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsafdsjhuidosahfudsah89320hfuepsahfijdsafdasfdsafdsafdsafdsafdasfdsafdsafdsfdsafdsafdsaf |
| 23:25:53 | raptor | ok, so i have a web-hook test script... |
| 23:26:02 | raptor | wasn't expecting it to work, honestly |
| 23:27:35 | kaen | I always wondered how you made logbot talk |
| 23:27:35 | sam686 | it might also be google code auto-disabling submittihg to bitfighter.org after its been inaccessable in over a few days? |
| 23:27:48 | kaen | ^ I was thinking the same thing |
| 23:28:23 | raptor | well, the web-hook also doesn't exist |
| 23:28:31 | raptor | looks like i just committed that test script, not the real one |
| 23:30:40 | raptor | in order to get google to work with a web hook, i had to do some sort of HMAC verification with a special key |
| 23:31:21 | raptor | then once verified, i had to deserialize the JSON and post the appropriate message to a socket i created on logbot |
| 23:31:30 | sam686 | I remember that one day, I telnet bflogbot as back then, it was wide open to anyone connecting to it |
| 23:31:30 | BFLogBot | no fortune here! |
| 23:31:43 | raptor | haha |
| 23:32:10 | raptor | well we must have fortune! |
| 23:33:12 | raptor | BFLogBot: give us wisdom |
| 23:33:12 | BFLogBot | The inside of my head was exploding with fireworks. Fortunately, my last thought turned out the lights when it left. -- Calvin |
| 23:33:22 | raptor | better |
| 23:38:46 | sam686 | My home server is just idling at about 0.1 CPU usage http://sam6.25u.com/cacti/graph.php?action=view&local_graph_id=2&rra_id=all |
| 23:39:09 | sam686 | I wonder what kaen's build bot server cpu usage at? |
| 23:39:21 | kaen | 200% |
| 23:39:27 | kaen | whenever a build is running |
| 23:39:33 | Watusimoto | >> GameType is more like what a Game class would be << Exactly! |
| 23:40:02 | raptor | does anyone else's brain break when you see a number like '200%' ? |
| 23:40:07 | sam686 | The big giant cpu usage in december was making my server generate heat as I was cold |
| 23:40:17 | kaen | lol sam |
| 23:40:35 | kaen | Watusimoto: I have some other commentary on the class hierarchy, if you dare to open that can of worms... |
| 23:40:58 | raptor | things raptor would like to do before another huge refactor: |
| 23:41:03 | Watusimoto | sure |
| 23:41:04 | raptor | 1. fix Level |
| 23:41:07 | Watusimoto | always interested |
| 23:41:18 | raptor | 2. fix Settings/GameSettings/IniSettings |
| 23:41:43 | kaen | I'll do a writeup and pastie it |
| 23:41:46 | raptor | or rather: things raptor would like to see get done :) |
| 23:45:55 | raptor | master doesn't compile now unless you do MASTER_MINIMAL - because there is something wrong with the Settings system |
| 23:48:16 | Watusimoto | my son's minecraft server is also out of comission... it feels like the internet is broken! |
| 23:49:01 | raptor | was that on the same server? |
| 23:49:14 | sam686 | or the same server provider? |
| 23:50:23 | Watusimoto | different provider |
| 23:50:36 | Watusimoto | in germany somewhere |
| 23:50:37 | Watusimoto | https://urpad.net/blog/update-627-evening/ |
| 23:50:45 | Watusimoto | latest from the bf host |
| 23:50:48 | Watusimoto | haven't read it yet |
| 23:51:23 | raptor | roughly: "12 hours, except for those we can't restore" |
| 23:51:28 | sam686 | https://urpad.net/blog/update-628-morning/ |
| 23:51:39 | Watusimoto | I tried to reboot ours a few minutes ago... no luck |
| 23:52:04 | Watusimoto | mmm... I have a bad feeling |
| 23:53:17 | Watusimoto | I'll open a new ticket tomorrow |
| 23:53:37 | Watusimoto | we're on LABL4, I think |
| 23:54:09 | Watusimoto | well, that's our "node" |
| 23:54:25 | Watusimoto | it does not match up nicely with the names referenced in the news |
| 23:55:09 | raptor | labl4 was not mentioned in the news, so that's good news... right? RIGHT? |
| 23:55:22 | Watusimoto | oh yeah, awesome news! |
| 23:55:58 | Watusimoto | latest push does not fix editor, just first assert |
| 23:56:04 | Watusimoto | don;t get your hopes up |
| 23:56:53 | sam686 | kaen can still connect to sam6.25u.com and someday move his build bot there if he doesn't have any other good place to run build bot |
| 23:58:12 | sam686 | besides bitfighter editor being broken, in-game forcefields all point to the right, going through walls |
| 23:59:05 | Watusimoto | what do we do about this? |
| 23:59:06 | Watusimoto | Error 1 error : Target bitfighter_client has dependency information when it shouldn't. C:\Users\Chris\Documents\bf-trunk\build\zap\CUSTOMBUILD bitfighter_client |
| 23:59:14 | Watusimoto | got after latest update |
| 23:59:21 | Watusimoto | I tried a full rebuild |
| 23:59:22 | raptor | did you re-run cmake? |
| 23:59:26 | Watusimoto | no |
| 23:59:29 | Watusimoto | ok, will try that |
| 23:59:35 | sam686 | delete cmakecache and run cmake? |
| 23:59:40 | raptor | yes, sorry, I thought I mentioned you'd have to |
| 23:59:45 | Watusimoto | you probably did |
| 23:59:49 | raptor | i made significant build system changes |
