Timestamps are in GMT/BST.
| 00:02:00 | | BFLogBot - Commit 955e910b07f2 | Author: buckyballreaction | Log: Minor hack to get Mac .app running again... |
| 00:09:00 | Watusimoto | ok, windows installer fixed, everything works |
| 00:10:00 | Watusimoto | crossing the icon off my list |
| 00:10:00 | karamazovapy | ey raptor |
| 00:10:00 | raptor | hi |
| 00:10:00 | karamazovapy | Lone Wolf sent me some 016 feedback |
| 00:10:00 | karamazovapy | apparently I'm your proxy |
| 00:10:00 | raptor | what |
| 00:11:00 | raptor | ok |
| 00:11:00 | karamazovapy | prepare for flood |
| 00:11:00 | raptor | PM me... |
| 00:12:00 | karamazovapy | I explained that #1 more or less always existed, that the reactor mode is still kind of under construction, and that the editor is likely still buggy, but that I'd pass it along |
| 00:12:00 | raptor | great! |
| 00:12:00 | raptor | thanks |
| 00:12:00 | karamazovapy | I assume the reactor mode isn't really "finished", anyway |
| 00:12:00 | | BFLogBot - Commit 809f0fc47ac2 | Author: watusim...@bitfighter.org | Log: Properly set window icon, and add icon to Windows installer. Need to get this into the Linux and Mac installers. And figure out why windows installs crash. |
| 00:13:00 | | BFLogBot - Commit e754691b7fcb | Author: watusim...@bitfighter.org | Log: Merge |
| 00:13:00 | raptor | oh karamazovapy, can you log in here and see if you can edit pages?: http://199.192.229.168/user |
| 00:13:00 | karamazovapy | the ship angle thing is just a matter of relative cursor position |
| 00:13:00 | raptor | Watusimoto: I figured out roles in drupal and gave omnipotency to karamazovapy, sam686 and myself for editing pages... |
| 00:13:00 | karamazovapy | won't let me log in |
| 00:14:00 | Watusimoto | great |
| 00:14:00 | karamazovapy | "Sorry, unrecognized username or password." |
| 00:14:00 | raptor | i see a user 'karamazovapy' |
| 00:14:00 | karamazovapy | I guess I can request a new password... |
| 00:15:00 | raptor | change it for you temporarily |
| 00:15:00 | raptor | seePM |
| 00:15:00 | Watusimoto | knocking off early tonight -- got to set out the St Nicholas Day presents |
| 00:15:00 | raptor | ha! |
| 00:15:00 | raptor | g'night |
| 00:15:00 | Watusimoto | just what my kids need -- more stuff |
| 00:15:00 | Watusimoto | ok, see you folks tomorrow |
| 00:16:00 | karamazovapy | fixed |
| 00:16:00 | raptor | see you |
| 00:16:00 | karamazovapy | later! |
| 00:16:00 | raptor | karamazovapy: you can get in now and edit pages? |
| 00:16:00 | raptor | this is the new bitfighter.org server bytheway |
| 00:16:00 | karamazovapy | well...I've logged in successfully |
| 00:16:00 | karamazovapy | yes, it looks like I can edit |
| 00:17:00 | raptor | great! |
| 00:18:00 | karamazovapy | it looks like comments aren't really "active", but somehow we've got some hidden away |
| 00:18:00 | karamazovapy | zusecon (not verified) Says: |
| 00:18:00 | karamazovapy | May 18, 2010 - 2:05pm |
| 00:18:00 | karamazovapy | sWZLER viagra 448 aciphex 504550 xanax rxp acomplia AR 8]] buy tramadol wholesale =-OOO ultram online unprescibed ndf buy ultram 002 |
| 00:18:00 | karamazovapy | � |
| 00:18:00 | karamazovapy | maybe just the one |
| 00:19:00 | raptor | haha, yeah i actually deleted about 50 spambots |
| 00:19:00 | karamazovapy | I suppose we're better off without comments enabled on the news page, since we have the forums |
| 00:19:00 | raptor | yes |
| 00:20:00 | raptor | registering for the main site gives you zero access... |
| 00:20:00 | raptor | i'm suprised someone got a comment in |
| 00:20:00 | karamazovapy | do we want to wait for 016 to update the screenshots? |
| 00:20:00 | raptor | nah |
| 00:20:00 | raptor | if you have some, change away! |
| 00:20:00 | karamazovapy | I don't really, they're just kind of outdated |
| 00:21:00 | karamazovapy | old loadout indicators and whatnot |
| 00:22:00 | raptor | I bet sam would have some good ones - if he ever comes back |
| 00:24:00 | karamazovapy | it'd be easy enough to take some with bots |
| 00:25:00 | raptor | those youtube vids would lots of good ones, too |
| 00:28:00 | | Watusimoto Quit (Ping timeout: 240 seconds) |
| 00:28:00 | karamazovapy | uh...what? |
| 00:32:00 | raptor | sorry - the youtube videos might have lots of good screenshots |
| 00:35:00 | karamazovapy | yeah...unfortunately the quality drops with compression |
| 00:36:00 | raptor | you'd need to choose a B-frame |
| 00:36:00 | raptor | sorry I-frame |
| 00:36:00 | raptor | but yeah - resolution isn't what we want anymore |
| 00:36:00 | raptor | i guess |
| 00:40:00 | karamazovapy | yeah, all my uncompressed stuff is gone |
| 00:59:00 | raptor | explain again what the commander's map pointing the wrong way is? |
| 00:59:00 | raptor | i don't see it... |
| 01:14:00 | | raptor Quit (Remote host closed the connection) |
| 01:28:00 | karamazovapy | raptor: if you read the logs... |
| 01:29:00 | karamazovapy | let's say your cursor is just left of center and has the ship aiming to the left hand side of the screen. |
| 01:32:00 | karamazovapy | now when you zoom out to the commander's map, since your own ship is no longer in the center of the screen as the focus of the map, it's possible to appear on the opposite side of your cursor |
| 01:33:00 | karamazovapy | because the cursor position determines your ship's heading, you can jump between facing full left in regular view to full right in command view without moving the mouse |
| 01:33:00 | karamazovapy | usually it's more subtle than that, but that's what he's talking about in regard to ship angle changing when you move in and out of command |
| 01:58:00 | | sam686 has joined |
| 01:58:00 | | ChanServ sets mode +v sam686 |
| 02:14:00 | | raptor has joined |
| 02:14:00 | | ChanServ sets mode +o raptor |
| 02:14:00 | raptor | thanks karamazovapy, i understand nw |
| 02:14:00 | raptor | sam686! you're back! |
| 02:15:00 | sam686 | hi |
| 02:15:00 | raptor | in your new place? |
| 02:15:00 | sam686 | yes |
| 02:17:00 | raptor | did you move out for school? |
| 02:17:00 | sam686 | no, but move to live in a different place, as the old house i was in can get a bit messy.. |
| 02:18:00 | raptor | ah... 'greener pastures' |
| 02:20:00 | raptor | any roommates? |
| 02:20:00 | sam686 | in a small apartment, i do have a roommate |
| 02:20:00 | raptor | cool |
| 02:20:00 | sam686 | but there is 2 bedrooms |
| 02:21:00 | raptor | well i'm glad you have internet access.. |
| 02:22:00 | sam686 | yes, except only through wireless internet for now, which probably means no access to my web page yet (not directly connected to internet) |
| 02:22:00 | sam686 | also, to get my desktop without wifi support to go to internet, i have to share my laptop's wireless connection to wire up to desktop.. |
| 02:23:00 | raptor | dinner time! |
| 02:24:00 | sam686 | although i seem to have good speed on this connection, same 5 Mbps upload, but slower download speed (about 16 Mbps, probably due to wireless G connection) |
| 02:58:00 | raptor | hi |
| 02:59:00 | sam686 | hi? |
| 02:59:00 | raptor | sam686: lone wolf mentioned a bug in windiws editor: resizing in windowed mode didn't resize the whole editor |
| 03:00:00 | raptor | does that happen for you? |
| 03:00:00 | raptor | in 016 |
| 03:01:00 | sam686 | what does it mean it doesn't resize the whole editor? |
| 03:01:00 | sam686 | it seems to work fine for me, unless you can show me some screenshots |
| 03:03:00 | raptor | m e too |
| 03:03:00 | raptor | in linux |
| 03:14:00 | sam686 | oh, could it be that a person holping it to resize to a non 4:3 ratio in windowed mode? |
| 03:15:00 | raptor | sent him an e-mail |
| 03:15:00 | raptor | not sure |
| 03:17:00 | raptor | what happens if you maximize the game? |
| 03:18:00 | raptor | is that what you;re talking about? |
| 03:18:00 | sam686 | probably yes |
| 03:19:00 | raptor | i think we need to solve a bug |
| 03:19:00 | | CrazyLinuxNerd has joined |
| 03:20:00 | raptor | remember that small level of mine that crashed thgame when trying to test it from the editor? |
| 03:22:00 | raptor | wait, i get a segfault with watusimoto's letest changes |
| 03:23:00 | raptor | #0 0x00000000005c39b1 in Zap::InitSdlVideo () at main.cpp:866 |
| 03:23:00 | raptor | #1 0x00000000005c4d96 in main (argc=1, argv=0x7fffffffde68) at main.cpp:1315 |
| 03:24:00 | raptor | ha! with the new BMP |
| 03:26:00 | raptor | haha, ok, solved it - it just needs to be handled better.... |
| 03:28:00 | raptor | also, he is mixing line endings... |
| 03:29:00 | | BFLogBot - Commit 3f0f7dd3a1a4 | Author: buckyballreaction | Log: Gracefully handle the new icon |
| 03:32:00 | raptor | ok now to that bug again.. |
| 03:34:00 | raptor | sam686: this level still crashes the game when trying to test it from the editor: http://www.maxhushahn.com/upload/test_soccer.level |
| 03:35:00 | | sam686 Quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/) |
| 03:41:00 | | sam686 has joined |
| 03:41:00 | | ChanServ sets mode +v sam686 |
| 03:44:00 | sam686 | now let me see what my quit message is, i think i changed it.. |
| 03:44:00 | | sam686 Quit (Quit: ?) |
| 03:45:00 | | sam686 has joined |
| 03:45:00 | | ChanServ sets mode +v sam686 |
| 03:46:00 | raptor | it was '?' |
| 03:46:00 | sam686 | yes, i couldn't think of what to say in quit message.. |
| 03:47:00 | CrazyLinuxNerd | sam686: Why not just leave it as default? |
| 03:47:00 | sam686 | what if i leave it blank? |
| 03:48:00 | sam686 | ok, a blank message comes, i think.. testing |
| 03:48:00 | | sam686 Quit (Client Quit) |
| 03:48:00 | raptor | i leave mine blank |
| 03:48:00 | CrazyLinuxNerd | yeah you can do that, but i was just curious as to why you would need to change it :-). |
| 03:48:00 | CrazyLinuxNerd | .. and his quit |
| 03:48:00 | | sam686 has joined |
| 03:48:00 | | ChanServ sets mode +v sam686 |
| 03:50:00 | sam686 | it gets rather annoying to see my own quit message in http://209.141.63.138/irclogs/ |
| 03:50:00 | CrazyLinuxNerd | stop quitting then xD |
| 03:51:00 | sam686 | well, sometimes i accidently close the KVirc, or i need to close it for some reason.. |
| 03:51:00 | raptor | sam686: notice something different?: http://199.192.229.168/ |
| 03:51:00 | raptor | :) |
| 03:51:00 | raptor | maybe we'll get more traffic... |
| 03:52:00 | sam686 | what is it, the message "No Bitfighter games are currently online. Why don't you host one?" |
| 03:52:00 | raptor | look above that |
| 03:53:00 | | testestblahlah has joined |
| 03:53:00 | sam686 | live chat? |
| 03:53:00 | testestblahlah | yep |
| 03:53:00 | | testestblahlah has left |
| 03:53:00 | raptor | :) |
| 03:54:00 | | CrazyLinuxNerd waits for people to starting asking "what is irc" |
| 03:54:00 | sam686 | can you try to avoid href="http://199.192.229.168/rss.xml" and just do href="/rss.xml" ? it may become a problem when the server is not 199.192.229.168 anymore |
| 03:55:00 | raptor | good addition? |
| 03:55:00 | sam686 | i can see that in "View source" |
| 03:55:00 | raptor | where is that? |
| 03:55:00 | sam686 | home page |
| 03:55:00 | | Test has joined |
| 03:55:00 | raptor | oh actually, that is dynamically generated by drupal |
| 03:56:00 | raptor | I checked a lot of those - they are actually stored as relative links in the DB... |
| 03:57:00 | raptor | hi Test |
| 03:57:00 | | Test Quit (Client Quit) |
| 04:03:00 | sam686 | that "live chat" may have a bug, if your nick is "." it fails to connect due to bad name |
| 04:04:00 | raptor | probably - it's just an <iframe> that freenode let's us use |
| 04:06:00 | raptor | i'm sending watusimoto an e-mail about that bug in the editor... |
| 04:07:00 | | sam686 Quit (Read error: Connection reset by peer) |
| 04:09:00 | karamazovapy | nice move, with Live Chat. |
| 04:09:00 | raptor | hi karamazovapy, good idea? |
| 04:09:00 | raptor | bad idea? |
| 04:09:00 | karamazovapy | good idea, I think |
| 04:09:00 | | sam686 has joined |
| 04:09:00 | | ChanServ sets mode +v sam686 |
| 04:10:00 | karamazovapy | I'm not sure homepage visitors will stick around long enough to receive a response, but it's a good thought |
| 04:10:00 | raptor | yeah... that's what i figure, too.. |
| 04:13:00 | raptor | but at least now we have access to the website! |
| 04:15:00 | sam686 | reCaptha sometimes have a problem with sometimes showing some very strange symbols like it happened to me. https://www.google.com/search?q=recaptha+symbols&tbm=isch |
| 04:17:00 | raptor | hahaha, yeah, i get those too |
| 04:18:00 | raptor | ok, sent an e-mail about that evil bug |
| 04:24:00 | raptor | sam686: lone wolf gave a screenshot of his error: http://www.maxhushahn.com/upload/error.png |
| 04:25:00 | karamazovapy | the 4chan response is to use the most offensive possible word as a substitute for the symbols |
| 04:27:00 | sam686 | that problem doesn't happen in 015a, but does happen in 016 |
| 04:27:00 | sam686 | i get it too |
| 04:28:00 | sam686 | its mostly because the window size always seem to reset to a small window size everytime you go from fullscreen to windowed, or when entering editor |
| 04:29:00 | raptor | how did you get it? |
| 04:29:00 | raptor | click the maximize button? |
| 04:30:00 | karamazovapy | try cycling between screen display options |
| 04:30:00 | karamazovapy | fullscreen stretched, fullscreen windowed, etc |
| 04:31:00 | raptor | when should i cycle? |
| 04:31:00 | raptor | still not getting it... |
| 04:31:00 | sam686 | just resize window, then enter editor, does it changes size by itself? |
| 04:32:00 | sam686 | it does for me |
| 04:32:00 | raptor | ah, it goes back to the original size |
| 04:32:00 | raptor | you're right... the scaling ratio isn't being saved... |
| 04:32:00 | sam686 | better yet, try maximizing window, and enter editor.. |
| 04:33:00 | raptor | i did - and it fills up to the 4:3 ratio |
| 04:33:00 | raptor | but maximizing in linux spikes the CPU, too... remember? |
| 04:33:00 | sam686 | maybe it should detect maximized, and not try to resize or move the window |
| 04:33:00 | sam686 | when maximized... |
| 04:36:00 | sam686 | i will need to go to bed, bye.. |
| 04:36:00 | raptor | night |
| 04:41:00 | | sam686 Quit (Ping timeout: 258 seconds) |
| 05:02:00 | raptor | ok ok... i'll try and work out some of the bugs... |
| 05:03:00 | raptor | you've convinced me |
| 05:17:00 | | Zoomber has joined |
| 05:17:00 | | ChanServ sets mode +v Zoomber |
| 05:20:00 | | Zoomber Quit (Client Quit) |
| 05:41:00 | | BFLogBot - Commit c7cc513b101a | Author: buckyballreaction | Log: Fix saving window scaling position. This also fixes an editor-maximize bug reported by Lone Wolf |
| 05:41:00 | raptor | great! now back to freecol... |
| 07:36:00 | | raptor Quit (Remote host closed the connection) |
| 08:20:00 | | kodax has joined |
| 12:46:00 | | CrazyLinuxNerd Quit (Quit: Leaving) |
| 12:59:00 | | CrazyLinuxNerd has joined |
| 15:57:00 | | watusimoto has joined |
| 15:57:00 | | ChanServ sets mode +o watusimoto |
| 16:20:00 | | raptor has joined |
| 16:20:00 | | ChanServ sets mode +o raptor |
| 16:20:00 | raptor | good day! |
| 16:22:00 | raptor | watusimoto: any luck getting bitfighter.net pointing to the new server? |
| 16:25:00 | raptor | someone on protocol 21 is playing... |
| 16:43:00 | watusimoto | what does that mean? |
| 16:43:00 | watusimoto | old game? |
| 16:44:00 | watusimoto | and no, no luck |
| 16:44:00 | watusimoto | I'm really unsure how to proceed |
| 16:44:00 | raptor | ok |
| 16:44:00 | raptor | yeah 21 = old game |
| 16:44:00 | watusimoto | just one player? |
| 16:44:00 | raptor | yep |
| 16:44:00 | watusimoto | sigh |
| 16:45:00 | watusimoto | I may try using my registrar's nameservers |
| 16:47:00 | raptor | DNS was always slightly convoluted to me... |
| 16:50:00 | watusimoto | taking a look at the problem listed in your email -- why does mTeams.clear(); cause a crash? |
| 16:50:00 | raptor | that's the thing - i never found the 'why' |
| 16:50:00 | watusimoto | right. I have some ideas |
| 16:50:00 | watusimoto | well, one idea |
| 16:51:00 | raptor | i think it has to do with boost trying to clean up one variation of Team when it should be cleaning up another... |
| 16:51:00 | raptor | but i'm just spouting off nonsense because i spent hours on the problem and couldn't figure it out at the time.. |
| 16:53:00 | watusimoto | gets called from mGame->clearTeams(); |
| 16:53:00 | watusimoto | what if mGame is NULL? |
| 16:54:00 | watusimoto | I'd add an assert in GAMETYPE_RPC_S2C(GameType, s2cAddTeam |
| 16:54:00 | watusimoto | First line: TNLAssert(mGame, "mGame is NULL!"); |
| 16:55:00 | watusimoto | not sure that's what's happening, but it's consistent |
| 16:56:00 | watusimoto | also, could try removing objects from level |
| 16:56:00 | raptor | i've done that |
| 16:56:00 | watusimoto | or adding name/descr |
| 16:56:00 | watusimoto | changing gametype |
| 16:56:00 | raptor | i've done all that |
| 16:56:00 | raptor | crashes |
| 16:56:00 | watusimoto | you posted a spare level, but not a minimal level |
| 16:57:00 | raptor | it has to do with the AbstractTeam and EditorTeam casting |
| 16:57:00 | watusimoto | sent, not posted |
| 16:57:00 | watusimoto | but why only on this level? |
| 16:57:00 | raptor | did you read the irc log url |
| 16:57:00 | watusimoto | no |
| 16:57:00 | watusimoto | :-) |
| 16:57:00 | raptor | not only on this level - on many minimal levels with just two teams it crashes for me |
| 16:57:00 | watusimoto | but, my best guess so far is the null mgame |
| 16:57:00 | raptor | it's not null |
| 16:57:00 | watusimoto | are you sure? |
| 16:57:00 | raptor | let me find that irc log again... |
| 16:58:00 | raptor | because it was a while ago, the log explains it better than i can right now |
| 16:58:00 | watusimoto | I've seen this kind of thing before, where the standard containers crash where they shouldn't |
| 16:58:00 | watusimoto | I'll read the log |
| 16:58:00 | raptor | i even narrowed it down to the exact line it crashes on in the log... |
| 16:59:00 | watusimoto | that's a long log :-) |
| 16:59:00 | raptor | search for Game::clearTeams() |
| 16:59:00 | watusimoto | 20:54 |
| 16:59:00 | raptor | yes |
| 17:00:00 | raptor | sam686 and I figured it was a memory corruption somewhere - even the pointers on windows were messed up, although it didn't crash |
| 17:00:00 | watusimoto | could t.get() be null? |
| 17:00:00 | raptor | i don't remember |
| 17:01:00 | watusimoto | still reading |
| 17:02:00 | raptor | sam686 kind of interjected another problem into the conversation... we seem to do that a lot.. |
| 17:02:00 | watusimoto | :-) |
| 17:02:00 | watusimoto | ok, so this problem recurs with many 2-team levels, right? |
| 17:03:00 | watusimoto | have you tried 3 team levels? |
| 17:03:00 | raptor | nope |
| 17:03:00 | watusimoto | or 1 team levels? |
| 17:03:00 | raptor | 1 team levels, yes |
| 17:03:00 | raptor | they seem to work find |
| 17:03:00 | raptor | fine |
| 17:03:00 | watusimoto | interesting |
| 17:03:00 | raptor | 2-team, many crash |
| 17:03:00 | watusimoto | ok |
| 17:04:00 | raptor | but some don't - it seems random, which is why i think it is a memory corruption somewhere |
| 17:04:00 | watusimoto | do you know if it crashes when the 2nd team is added? |
| 17:04:00 | watusimoto | (teams are added 1 by 1) |
| 17:04:00 | raptor | you mean start a level with one team, test -> works, then go back, add 2nd team -> crash ? |
| 17:04:00 | watusimoto | log implies the answer is yes |
| 17:05:00 | watusimoto | no, I mean when 2 teams, first team 1 is sent from server to client then team 2 |
| 17:05:00 | watusimoto | my question is if you know if the crash happens when the first team is sent, ro the second |
| 17:05:00 | watusimoto | you wrote: on level test, team 2 (the one that crashes) |
| 17:05:00 | watusimoto | suggesting the answer is 2 |
| 17:05:00 | raptor | 2nd team |
| 17:05:00 | watusimoto | which makes sense if 1 team doesn't crash |
| 17:05:00 | raptor | it always seem to work on the first for me |
| 17:05:00 | watusimoto | and implying that 3 teams would crash also |
| 17:05:00 | watusimoto | so. |
| 17:06:00 | watusimoto | the problem is probably not mgame == null |
| 17:06:00 | raptor | crash ocurred when casting the second team... |
| 17:06:00 | watusimoto | when casting the 2nd team, or clearing the array that had the first in it? |
| 17:06:00 | watusimoto | not array, vector |
| 17:06:00 | watusimoto | the stack trace suggests it's the clear that's the problem |
| 17:07:00 | watusimoto | which may mean something is wrong with a destructor, or something has been cast to something it shouldn't have been |
| 17:07:00 | raptor | i can't answer definitively - because the stack trace was completely different on sam's machine |
| 17:07:00 | watusimoto | that's fine, just thinking out loud |
| 17:08:00 | watusimoto | ok, I've read the log from the line you mentioed to the end; I have to go home; will be back on later (if I don't fall asleep), and I'll work on this problem tonight. |
| 17:08:00 | raptor | ok |
| 17:08:00 | raptor | thanks for reading the log.. :) |
| 17:09:00 | watusimoto | not sure how to test it without linux handy, but I'll at least try to get a better idea of what's going on |
| 17:09:00 | watusimoto | I understand your frustration :-) |
| 17:09:00 | | Watusimoto_ has joined |
| 17:10:00 | watusimoto | or |
| 17:10:00 | watusimoto | or maybe boost is getting rid of something that had already been deleted elsewhere somehow |
| 17:10:00 | watusimoto | that's my current best hypothesis |
| 17:10:00 | raptor | looks like someone turned on your laptop at home.. |
| 17:10:00 | raptor | yes, that could be it |
| 17:11:00 | watusimoto | indeed it does |
| 17:11:00 | watusimoto | my kids have the day off today, maybe playing computer games? |
| 17:11:00 | watusimoto | here is is st. nicholas day |
| 17:11:00 | watusimoto | a school holiday |
| 17:11:00 | raptor | you should ssh in and play random songs from the command line |
| 17:11:00 | watusimoto | presents |
| 17:11:00 | watusimoto | it's windows |
| 17:11:00 | raptor | awww |
| 17:12:00 | watusimoto | very last line of stacktrace: |
| 17:12:00 | watusimoto | Program received signal SIGSEGV, Segmentation fault. 0x00000000004146ea in boost::detail::sp_counted_base::release (this=0x10574f0) at ../boost/boost/smart_ptr/detail/sp_counted_base_gcc_x86.hpp:145 145 dispose(); |
| 17:12:00 | raptor | yeah, trying to get rid of something that's already gone... |
| 17:12:00 | watusimoto | it's just when it tries to unload the memory for the object you just deleted |
| 17:12:00 | watusimoto | yup, I'll bet that's it |
| 17:12:00 | watusimoto | I'll poke around tonight and see if I can see it. |
| 17:12:00 | raptor | ok |
| 17:12:00 | watusimoto | later |
| 17:13:00 | raptor | bye |
| 17:13:00 | raptor | Watusimoto_: we know what you're doing... |
| 17:13:00 | raptor | :-) |
| 17:23:00 | | CrazyLinuxNerd Quit (Ping timeout: 240 seconds) |
| 17:23:00 | | CrazyLinuxNerd has joined |
| 17:23:00 | | CrazyLinuxNerd Quit (Changing host) |
| 17:23:00 | | CrazyLinuxNerd has joined |
| 17:33:00 | | kodax Quit (Remote host closed the connection) |
| 17:45:00 | | Watusimoto_ Quit (Ping timeout: 248 seconds) |
| 18:33:00 | | LordDVG has joined |
| 20:04:00 | | CrazyLinuxNerd Quit (Quit: Leaving) |
| 20:31:00 | | Watusimoto_ has joined |
| 20:31:00 | Watusimoto_ | hi |
| 20:31:00 | raptor | hi |
| 20:31:00 | Watusimoto_ | on to your special bug |
| 20:32:00 | raptor | if you think you found something, I can apply a patch and test it |
| 20:33:00 | Watusimoto_ | ok, will do. I'm going to start by assuming that though windows doesn;t crash, we have the same problem |
| 20:34:00 | raptor | i've wondered if linux is more aggressive in freeing memory than windows |
| 20:35:00 | Watusimoto_ | now I'm pretty sure we don't double delete the object |
| 20:35:00 | Watusimoto_ | after reading through a little more code |
| 20:35:00 | Watusimoto_ | phooey |
| 20:37:00 | Watusimoto_ | but I'll put a breakpoint on the destructor just ot be sure |
| 20:37:00 | Watusimoto_ | and a couple of asserts here and there |
| 20:38:00 | raptor | if only I could debug as fast I as can refactor with eclipse... |
| 20:41:00 | Watusimoto_ | no kidding... it's a hugely useful tool |
| 20:41:00 | Watusimoto_ | learn gdb |
| 20:42:00 | Watusimoto_ | there is a graphical front end, so I'm told |
| 20:42:00 | Watusimoto_ | I know how you hate the cmd line |
| 20:43:00 | | LordDVG Quit (Remote host closed the connection) |
| 20:44:00 | raptor | i learned gdb |
| 20:45:00 | raptor | the 'graphical' front-end is 'ddd' |
| 20:46:00 | raptor | it's just being able to traverse variables several levels deep with the click of a mouse is much nicer... |
| 20:46:00 | raptor | i'm spoiled with java... |
| 20:50:00 | karamazovapy | http://bitfighter.org/forums/viewtopic.php?f=9&t=1039&p=8691#p8691 |
| 20:53:00 | Watusimoto_ | these errors happen while testing from the editor, right? |
| 20:53:00 | Watusimoto_ | off topic |
| 20:53:00 | Watusimoto_ | blech |
| 20:54:00 | | koda has joined |
| 20:54:00 | karamazovapy | it's easy to hate off-topic, but it keeps the younger players engaged in the community. |
| 20:55:00 | Watusimoto_ | yes, I've come to realize that |
| 20:55:00 | karamazovapy | the other bonus is that it's a stress release valve for crap showing up in general discussion |
| 20:56:00 | Watusimoto_ | I've concluded that as well |
| 20:56:00 | Watusimoto_ | was just thinking that as I read the thread |
| 20:57:00 | Watusimoto_ | raptor |
| 20:57:00 | raptor | Watusimoto_: errors happen when testing from editor |
| 20:58:00 | Watusimoto_ | I've found no evidence of double deletes |
| 20:58:00 | raptor | loading the level directly from 'host game' works with no problems |
| 20:58:00 | Watusimoto_ | mgame is not null (as you stated earlier) |
| 20:58:00 | Watusimoto_ | host game works??? |
| 20:58:00 | raptor | yes |
| 20:58:00 | Watusimoto_ | so it's only transitioning btwn the editor and the game... |
| 20:59:00 | Watusimoto_ | we know it is a client side issue |
| 20:59:00 | raptor | that's why i think it's something goofy with recasting between EditorTeam and AbstractTeam and Team |
| 20:59:00 | Watusimoto_ | and client game may share team list with editor |
| 20:59:00 | Watusimoto_ | which does suggest the possiblity you stated |
| 21:00:00 | Watusimoto_ | ok, restart->start editor->test level->there are two teams already in cleintGame team list |
| 21:01:00 | raptor | yes |
| 21:01:00 | Watusimoto_ | that is the point it fails for you? |
| 21:01:00 | Watusimoto_ | not after restarting the level or letting the clock run out |
| 21:01:00 | Watusimoto_ | but when you first test? |
| 21:02:00 | raptor | as soon as i select 'test level' - crash! |
| 21:02:00 | Watusimoto_ | ok, then it is when the game deletes the editor's teams |
| 21:03:00 | Watusimoto_ | so I'm almost at the point where you already are |
| 21:03:00 | Watusimoto_ | Vector<boost::shared_ptr<AbstractTeam> > mTeams; |
| 21:05:00 | Watusimoto_ | there's nothing wrong with |
| 21:05:00 | Watusimoto_ | boost::shared_ptr<BaseClass> pbase(new DerivedClass()); |
| 21:05:00 | Watusimoto_ | according to stack overflow |
| 21:05:00 | Watusimoto_ | that's what we're doing |
| 21:05:00 | Watusimoto_ | but... |
| 21:05:00 | raptor | nope, boost can handle that |
| 21:05:00 | Watusimoto_ | there's an important caveat |
| 21:05:00 | Watusimoto_ | which I'm going to check |
| 21:05:00 | Watusimoto_ | then tell you |
| 21:05:00 | Watusimoto_ | (though you probably already know) |
| 21:06:00 | Watusimoto_ | caveat --> (Assuming in both cases that BaseClass has a virtual destructor.) |
| 21:06:00 | Watusimoto_ | but it appears all classes have virtual destructors. |
| 21:06:00 | Watusimoto_ | rats |
| 21:06:00 | raptor | evil virtual distructors - i've been cleaning those up with my header clean-up project... |
| 21:07:00 | raptor | ok, so i just tried something: |
| 21:07:00 | raptor | comment out UIEditor.cpp:471 and it doesn't crash anymore |
| 21:07:00 | raptor | // game->clearTeams(); |
| 21:08:00 | Watusimoto_ | yes, this is certainly the problem |
| 21:08:00 | Watusimoto_ | wait... uieditor? |
| 21:08:00 | raptor | oh man, but then the next level i load has 4 teams in it... |
| 21:09:00 | Watusimoto_ | and... |
| 21:09:00 | Watusimoto_ | what?!? |
| 21:09:00 | raptor | then the next has 6 |
| 21:09:00 | raptor | then 8 |
| 21:09:00 | Watusimoto_ | but no crashes |
| 21:09:00 | Watusimoto_ | oh, I see |
| 21:09:00 | raptor | got a crash |
| 21:09:00 | Watusimoto_ | yes, ok |
| 21:09:00 | raptor | let me duplicate |
| 21:09:00 | Watusimoto_ | that's to be expected |
| 21:09:00 | Watusimoto_ | I think |
| 21:09:00 | raptor | so that is a wild goose chase methings |
| 21:09:00 | Watusimoto_ | try removing this line: |
| 21:11:00 | Watusimoto_ | gametype.cpp 2168 |
| 21:11:00 | Watusimoto_ | if(firstTeam) |
| 21:11:00 | Watusimoto_ | mGame->clearTeams(); |
| 21:12:00 | raptor | those two? |
| 21:13:00 | Watusimoto_ | sure |
| 21:13:00 | Watusimoto_ | you will likely get the 2-4-6-8 teams |
| 21:13:00 | raptor | still getting the problem with 2 teams added each time i restart |
| 21:13:00 | | sam686 has joined |
| 21:13:00 | | ChanServ sets mode +v sam686 |
| 21:13:00 | raptor | crash! |
| 21:14:00 | Watusimoto_ | but crash when you have tons of teams |
| 21:14:00 | raptor | crashes when exiting editor and loading second level |
| 21:14:00 | Watusimoto_ | exiting editor and loading 2nd level? |
| 21:14:00 | raptor | yes |
| 21:15:00 | raptor | but i can't duplicated it when in gdb... |
| 21:15:00 | raptor | there we go |
| 21:15:00 | raptor | duplicated |
| 21:15:00 | raptor | backtrace: http://pastie.org/2977063 |
| 21:16:00 | raptor | crashes about %80 of the time if i: open level, test, exit editor, open new level, test, exit editor, open new level -> crash! |
| 21:17:00 | Watusimoto_ | ok, totally expected |
| 21:17:00 | raptor | with those two lines commented out - same behaviour when i remove clearTeams() from UIEditor::loadLevel |
| 21:17:00 | Watusimoto_ | too many levels makes things go bad |
| 21:17:00 | raptor | ok |
| 21:17:00 | Watusimoto_ | I don't understand why the UIEditor::loadLevel behaves that way |
| 21:18:00 | Watusimoto_ | as that should have nothing to do with whether teams in-game are deleted |
| 21:18:00 | Watusimoto_ | but I think I do understand what's happening |
| 21:18:00 | Watusimoto_ | and I think you are right about the deleting of team/editorteam objects |
| 21:18:00 | Watusimoto_ | not sure why though |
| 21:18:00 | Watusimoto_ | we can fix with a hack; create a different team list for the editor |
| 21:19:00 | raptor | i still don't think I understand, but I am willing to do what it takes for it to go away :) |
| 21:19:00 | Watusimoto_ | though... with that stack trace... what is the team editor doing while you are testing? |
| 21:19:00 | Watusimoto_ | it should not be doing anything |
| 21:19:00 | Watusimoto_ | why would it crash??? |
| 21:20:00 | Watusimoto_ | probably not importnat, but puzzling |
| 21:20:00 | raptor | i don't know - it crashed upon level open in editor |
| 21:20:00 | raptor | maybe something is a child of something it shouldn't be? |
| 21:21:00 | raptor | i've found stack traces can go all over the place if it is a memory corruption... |
| 21:21:00 | raptor | so i wouldn't rely on it too much |
| 21:23:00 | sam686 | try doing, enter editor, test level, return to editor, and try to edit team names |
| 21:23:00 | Watusimoto_ | oh |
| 21:24:00 | sam686 | problem is editor and testing share the same ClientGame team list |
| 21:24:00 | Watusimoto_ | I think it will crash before he can do that cycle |
| 21:24:00 | Watusimoto_ | yes, I have come to thtat conclusion as well |
| 21:24:00 | raptor | hi sam686 |
| 21:24:00 | Watusimoto_ | but there is no reason for that to be a problem |
| 21:24:00 | Watusimoto_ | hey sam, writing from your new apartment? |
| 21:24:00 | sam686 | yes |
| 21:24:00 | raptor | maybe boost is trying to clean up the object too aggressively? |
| 21:24:00 | sam686 | at the new place.. |
| 21:25:00 | raptor | or gcc doesn't see them as a similar type despite the virtual destructor... |
| 21:25:00 | raptor | could be a million reasons |
| 21:25:00 | raptor | ok maybe a dozen... |
| 21:27:00 | Watusimoto_ | ok |
| 21:27:00 | Watusimoto_ | fine |
| 21:28:00 | Watusimoto_ | I'll rewrite the team stuff without a shared vector |
| 21:28:00 | Watusimoto_ | give me 20 minutes |
| 21:28:00 | raptor | when in doubt split it out! |
| 21:28:00 | raptor | i can test a patch before you commit... |
| 21:28:00 | raptor | i don't know how to make nice patches in windows, though |
| 21:28:00 | raptor | sam686 probably does |
| 21:29:00 | raptor | oh sam686, please test that editor-not-resizing bug in windows now... i think I fixed it |
| 21:30:00 | sam686 | seems to work fine, i don't really see any sizing problems, |
| 21:31:00 | raptor | great! |
| 21:37:00 | Watusimoto_ | this is complicated... |
| 21:57:00 | Watusimoto_ | all kinds of duplicate code popping up... |
| 22:01:00 | raptor | as in - you are creating it? |
| 22:07:00 | Watusimoto_ | yes |
| 22:07:00 | Watusimoto_ | recreating it, actually |
| 22:07:00 | Watusimoto_ | since I combined the team stuff to get rid of it |
| 22:07:00 | raptor | :-/ |
| 22:08:00 | Watusimoto_ | mostly little 1-2 ine helper methods |
| 22:11:00 | sam686 | Could run editor as a seperate ClientGame, but not sure if that will work... |
| 22:11:00 | raptor | could this have something to do with it?: http://stackoverflow.com/questions/4195691/shared-ptr-to-use-shared-ptr-is-it-safe |
| 22:13:00 | Watusimoto_ | we don't do the casting |
| 22:13:00 | Watusimoto_ | I don't think that's our problem |
| 22:13:00 | raptor | ok |
| 22:13:00 | raptor | i wasn't sure if there was some casting somewhere going on... |
| 22:13:00 | Watusimoto_ | no |
| 22:13:00 | Watusimoto_ | but... don't feel too too bad |
| 22:14:00 | Watusimoto_ | it's a little tiny bit hacky to jam both team lists into the same object |
| 22:14:00 | raptor | so i can feel good about forcing you to not be a tiny bit hacky? |
| 22:16:00 | Watusimoto_ | a tiny bit happy |
| 22:17:00 | | sam686 Quit (Ping timeout: 258 seconds) |
| 22:18:00 | | sam686 has joined |
| 22:18:00 | | ChanServ sets mode +v sam686 |
| 22:23:00 | | sam686 Quit (Disconnected by services) |
| 22:24:00 | | sam686 has joined |
| 22:24:00 | | ChanServ sets mode +v sam686 |
| 23:03:00 | Watusimoto_ | falling asleep.. want to get this working |
| 23:03:00 | raptor | wake up earlier to do it? |
| 23:03:00 | raptor | :) |
| 23:03:00 | Watusimoto_ | unlikely |
| 23:07:00 | Watusimoto_ | cannot convert parameter 1 from 'const TNL::Vector<T> *' to 'const TNL::Vector<T> *' |
| 23:07:00 | Watusimoto_ | not really sure what to make of this error |
| 23:07:00 | raptor | haha |
| 23:08:00 | Watusimoto_ | I know I'm tired,but it just doesn't make sense |
| 23:09:00 | raptor | sounds like an array issue? |
| 23:09:00 | raptor | usually i get that when trying to convert char[] to char* |
| 23:09:00 | raptor | but it is at least different on both sides of the 'to' |
| 23:11:00 | sam686 | cannot convert parameter 1 from 'const TNL::Vector<T> *' to 'const TNL::Vector<T> *' - usually looking at output log (instead of error list) will tell you what T equals |
| 23:18:00 | Watusimoto_ | I think I found the problem |
| 23:18:00 | Watusimoto_ | when loading levels, we run this: |
| 23:18:00 | Watusimoto_ | void Game::onReadTeamParam |
| 23:19:00 | Watusimoto_ | this contains the following code |
| 23:19:00 | Watusimoto_ | boost::shared_ptr<Team> team = boost::shared_ptr<Team>(new Team()); |
| 23:19:00 | Watusimoto_ | if(team->processArguments(argc, argv)) |
| 23:19:00 | Watusimoto_ | addTeam(team); |
| 23:19:00 | Watusimoto_ | i.e. it creates a Team |
| 23:19:00 | Watusimoto_ | but the editor deals with TeamEditors |
| 23:19:00 | raptor | ok |
| 23:19:00 | Watusimoto_ | and I think this may be the source of the problem |
| 23:20:00 | raptor | i'm in game.cpp:965 |
| 23:20:00 | raptor | and i see different code |
| 23:20:00 | Watusimoto_ | ah, what do you see |
| 23:20:00 | raptor | void Game::onReadTeamParam(S32 argc, const char **argv) |
| 23:20:00 | raptor | { |
| 23:20:00 | raptor | if(getTeamCount() < GameType::MAX_TEAMS) // Too many teams? |
| 23:20:00 | raptor | { |
| 23:20:00 | raptor | boost::shared_ptr<AbstractTeam> team = getNewTeam(); |
| 23:20:00 | raptor | if(team->processArguments(argc, argv)) |
| 23:20:00 | raptor | addTeam(team); |
| 23:20:00 | raptor | } |
| 23:20:00 | raptor | } |
| 23:21:00 | Watusimoto_ | oh |
| 23:21:00 | Watusimoto_ | ok then |
| 23:21:00 | Watusimoto_ | forget it |
| 23:22:00 | raptor | ? |
| 23:22:00 | raptor | so that is being overriden elsewhere? |
| 23:23:00 | Watusimoto_ | no, I'm seeing a modified version |
| 23:23:00 | Watusimoto_ | that was probably right, probably not the problem |
| 23:44:00 | | koda Quit (Quit: [koda release];) |
| 23:53:00 | Watusimoto_ | this is becoming a real mess |
| 23:59:00 | raptor | yikes |
| 23:59:00 | raptor | i should test whether this is a problem on mac... |
