| 00:00.05 | blast007 | mdskpr_: think I almost got a fix ready to go |
| 00:01.15 | CIA-114 | BZFlag: 03blast007 * r21506 10/trunk/bzflag/src/bzfs/bzfs.cxx: Get rid of some unused and shadowed variables |
| 00:01.17 | blast007 | mdskpr_: give that a shot |
| 00:01.58 | blast007 | if there's any in the client, I can bring up a VM (I was building this on my headless server, so no client) |
| 00:02.18 | mdskpr_ | k |
| 00:03.16 | delusional | i saw those unused vars too,,, also, a few in trackmarks.cxx http://pastebin.com/Rnndy5uu |
| 00:03.36 | mdskpr_ | ok that worked blast007 |
| 00:03.48 | mdskpr_ | but it may have been because i removed --enable-debug : / |
| 00:03.55 | blast007 | yeah |
| 00:03.58 | blast007 | best to leave that on |
| 00:04.04 | blast007 | since that treats warnings as errors |
| 00:04.47 | blast007 | delusional: I'll try the client build here |
| 00:07.44 | mdskpr_ | it worked blast007 |
| 00:08.03 | blast007 | with a clean build? |
| 00:08.07 | mdskpr_ | yes |
| 00:08.12 | blast007 | ok |
| 00:08.27 | blast007 | I'll build here too just to make sure |
| 00:09.07 | blast007 | see if I can get delusional's warning |
| 00:25.22 | blast007 | worked fine here too.. |
| 00:30.20 | McSpider | https://sourceforge.net/tracker/?func=detail&aid=3303071&group_id=3248&atid=303248 |
| 00:32.21 | *** join/#bzflag trepan (~trepan@unaffiliated/trepan) |
| 00:32.21 | *** mode/#bzflag [+v trepan] by ChanServ |
| 00:55.19 | Thumper_ | JeffM: I can't connect to servers anymore after r21501 http://www.norang.ca/tmp/r21501 |
| 00:55.32 | *** join/#bzflag catay (~smertens@kaiya.catay.be) |
| 01:30.25 | *** join/#bzflag meem1029 (~meem1029@2001:468:1930:208:5642:49ff:fe62:fe1d) |
| 01:33.28 | JeffM | Thumper_, what does it say? |
| 01:33.46 | Thumper_ | the server crashes, Error connecting to server. in the client |
| 01:33.55 | Thumper_ | trying a local server |
| 01:34.06 | Thumper_ | same on the public ones at 21506 |
| 01:34.09 | JeffM | well if the server crashes that's why it can't connect ;) |
| 01:34.10 | Thumper_ | (which are all dead now I think) |
| 01:34.14 | JeffM | run it in a debuger |
| 01:34.16 | Thumper_ | :) |
| 01:34.19 | *** part/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 01:34.44 | JeffM | I did test it and I connected to my local server |
| 01:36.35 | JeffM | linux may have problems with it if ares is failing, the first thing we do on connection is use ares to lookup the host, and I did change that around a little |
| 01:37.01 | JeffM | have to start that a little later then we used to cus we only want to do that for real players, not net connected peers |
| 01:37.05 | mdskpr_ | JeffM: #0 0x0000000000559724 in PlayerInfo::isPlaying (this=0x0) at ../../include/PlayerInfo.h:216 |
| 01:37.05 | mdskpr_ | #1 0x0000000000618c8b in NetHandler::~NetHandler (this=0x9feac0, __in_chrg=<value optimized out>) at NetHandler.cxx:332 |
| 01:37.05 | mdskpr_ | #2 0x00000000005b3293 in main (argc=3, argv=0x7fffffffdec8) at bzfs.cxx:5954 |
| 01:37.20 | JeffM | I have no idea what that says |
| 01:37.30 | blast007 | JeffM: if I'm going to add an enum to the BZFS API for the message type (chat or action), can the API access stuff from global.h? I notice that there's a team enum (with slightly different items/order) in bzfsAPI.h compared to what's in global.h |
| 01:37.41 | JeffM | blast007, nope |
| 01:37.48 | JeffM | that ONE header only |
| 01:37.48 | blast007 | ok |
| 01:37.54 | JeffM | thats why we copy them |
| 01:38.04 | JeffM | also helps if we change the guts then plugins don't have to change |
| 01:38.33 | JeffM | why an enum for just 2 states, use a bool? |
| 01:38.42 | JeffM | you think there will be other states later? |
| 01:39.32 | blast007 | perhaps |
| 01:39.36 | JeffM | k |
| 01:39.39 | blast007 | I'm sending the type as a byte anyway |
| 01:39.42 | JeffM | k |
| 01:40.44 | blast007 | should I be passing in the type to functions as a uint8_t or an int? seems like the playerIndex is sent in as an int in those functions |
| 01:41.06 | JeffM | it is an int because I want to have more then 256 players later |
| 01:41.23 | JeffM | for enums pass it in as the enum |
| 01:41.38 | blast007 | ok |
| 01:41.41 | JeffM | where are you adding it? |
| 01:41.47 | JeffM | cus I'm going to do the API rework next |
| 01:42.33 | blast007 | I might not touch the API initially |
| 01:42.37 | blast007 | I hadn't yet in 2.99.x |
| 01:42.57 | blast007 | though I probably won't finish tonight anyway |
| 01:43.04 | JeffM | if you want me to I can add what you need when I do the cleanup, just write up what you need |
| 01:43.37 | blast007 | so far I just have the enum added to the API |
| 01:43.55 | CIA-114 | BZFlag: 03JeffM2501 * r21507 10/trunk/bzflag/src/game/NetHandler.cxx: check for null player infos before we try to see if they are playing |
| 01:44.10 | JeffM | I have to ponder how I want to change the API, for better or for awesome |
| 01:50.38 | blast007 | hmm. For "BZF_API bool bz_sendTextMessagef(int from, int to, const char* fmt, ...);" I can't include the type with a default value, can I.. |
| 01:51.08 | JeffM | not on a ... one |
| 01:51.19 | JeffM | make 2 functions, one for chat and one for action |
| 01:51.34 | JeffM | make text message be the chat |
| 01:51.41 | JeffM | then add bz_sentTextAction |
| 01:51.49 | blast007 | k |
| 01:51.54 | JeffM | then have them call a common function in the api cpp file |
| 01:51.59 | blast007 | or ActionMessage ? |
| 01:52.20 | JeffM | we have moving stuff on screen, I' consider those actions, and they take network messages |
| 01:52.27 | JeffM | you are sending text that has the "action" property |
| 01:52.30 | blast007 | k |
| 01:52.34 | JeffM | so I'd have the word text in there to be clear |
| 01:54.09 | CIA-114 | BZFlag: 03JeffM2501 * r21508 10/trunk/bzflag/MSVC/build/ (bzadmin.sln bzflag.sln): since we have fullbuild we can make a solution for each exe to make debugging multiple apps ( i.e. server AND client ) simpler. |
| 01:57.13 | Thumper_ | bzflag3 trunk servers now check for commits every 15m and rebuild if there is something new - they'll pick up protocol changes faster now |
| 01:57.23 | JeffM | cool |
| 01:59.17 | JeffM | so..... how do people feel about the allman (AKA ISO C) style of indenting? :) |
| 02:02.26 | Thumper_ | I don't really care as long as we're consistent, it's documented in DEVINFO, and we're not changing styles regularly ;) I'm used to the existing style but meh. |
| 02:03.26 | JeffM | I can not stand the current style :) |
| 02:03.47 | JeffM | I can deal with the indent sizes but having { on the same line as the if just screws me up |
| 02:03.51 | trepan | much prefers BSD KNF to Allman |
| 02:04.09 | JeffM | yeah but you type in a 24 lin terminal on a greenscreen ;) |
| 02:04.39 | trepan | I'll have you know that I use 100x50 terms :) |
| 02:04.44 | JeffM | heh |
| 02:04.54 | trepan | 5 per screen, big font |
| 02:05.28 | trepan | 80 chars max, another good rule of thumb |
| 02:05.55 | JeffM | get larger monitors, widescreen FTW |
| 02:06.03 | trepan | 24", 1920 |
| 02:06.14 | JeffM | then you should have no width problems |
| 02:06.42 | trepan | humans read best vertically, break up the lines |
| 02:07.03 | JeffM | I always thought you were some kind of coding bot |
| 02:07.18 | trepan | is just very strict with his style |
| 02:08.24 | trepan | except for '} else {' vs. '}\n\telse {' -- oddly, I find that that one a matter of taste depending on the surrounding code :) |
| 02:08.36 | trepan | s/a matter/is a matter/ |
| 02:08.45 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 02:08.45 | *** mode/#bzflag [+v spldart] by ChanServ |
| 02:08.52 | JeffM | I like seeing a nearly blank line seperating my blocks |
| 02:10.10 | trepan | iirc, the 'go' language enforces something akind to BSD KNF; you can't have lines with just '{' |
| 02:10.27 | trepan | s/akind/akin/ |
| 02:11.51 | trepan | ya, the section on semicolons mentions it: http://golang.org/doc/go_tutorial.html |
| 02:12.13 | JeffM | ummm.. good for that language ... that we aren't using .... |
| 02:12.45 | trepan | i'm saying that that style is becoming more and more popular from what I'm seeing |
| 02:13.07 | JeffM | iso standard C code samples are all published in allman |
| 02:13.39 | trepan | a _lot_ of actual code is BSD KNF |
| 02:14.38 | JeffM | BulletCatcher, just FYI, when you add someone to the project make sure you add them to the patch tracker so they can be assigned tasks :) |
| 02:17.34 | McSpider | Someone mind checking the two patches I posted on the sf tracker? |
| 02:17.56 | JeffM | that's why I sent that message to BulletCatcher I'm trying to assign them to mdskiper |
| 02:18.17 | JeffM | cus yer just takin work off my plate |
| 02:18.42 | McSpider | Is that bad? |
| 02:18.47 | JeffM | oh god no |
| 02:18.51 | JeffM | I have the bigest plate |
| 02:18.56 | McSpider | good :) |
| 02:19.14 | JeffM | mdskiper has the smallest plate |
| 02:19.18 | JeffM | and he can apply patches |
| 02:19.37 | JeffM | I get an email for every patch someone adds for some reason. |
| 02:20.09 | JeffM | so I'll use him as patch monkey :) |
| 02:21.56 | trepan | i'll take the -publickey backport (just looked at the spreadsheet) |
| 02:22.13 | JeffM | k |
| 02:22.41 | trepan | really don't know if i'll have enough free time for the lua backport |
| 02:22.46 | JeffM | k |
| 02:23.02 | trepan | although if i don't do that, i will do the dynamic color BZDB tie-in backport |
| 02:23.13 | JeffM | makes sense |
| 02:23.28 | *** join/#bzflag catay (~smertens@kaiya.catay.be) |
| 02:24.23 | CIA-114 | BZFlag: 03JeffM2501 07http://my.bzflag.org * r7655 10/w/BZFlag_2.3: /* Backports */ |
| 02:26.46 | CIA-114 | BZFlag: 03JeffM2501 * r21509 10/trunk/bzflag/MSVC/build/ (3D.vcproj bzflag.vcproj game.vcproj): remove /Wp64 cus they are just called "warnings" now. |
| 02:27.33 | CIA-114 | BZFlag: 03Thumper 07http://my.bzflag.org * r7656 10/w/BZFlag_2.3: /* Backports */ |
| 02:35.24 | CIA-114 | BZFlag: 03JeffM2501 * r21510 10/trunk/bzflag/src/bzflag/ (BaseLocalPlayer.cxx BaseLocalPlayer.h playing.h): force the tank update to break the DT checks into chunks for long update times, like when we are taking a screen shot or doing another blocking op. make it update like the game was running at around 30fps. |
| 02:38.56 | JeffM | Thumper_, I'm going to try to get server side players in when I do the API update |
| 02:39.05 | Thumper_ | ok |
| 02:39.20 | JeffM | it should just be a copy and paste |
| 02:39.42 | JeffM | and some changes to gamekeeper |
| 03:26.27 | *** join/#bzflag Djpenguin (~David@c-98-207-67-102.hsd1.ca.comcast.net) |
| 04:22.49 | CIA-114 | BZFlag: 03trepan * r21511 10/trunk/bzflag/src/bzfs/ (4 files): * backported the -publickey bzfs option |
| 04:23.05 | trepan | also updated bzfls.php on the server |
| 04:25.14 | CIA-114 | BZFlag: 03trepan * r21512 10/trunk/db/bzfls.php: * reenabled the -publickey check |
| 04:37.11 | trepan | "+JeffM> iso standard C code samples are all published in allman" -- I'm looking at the ISO C99 spec, and the examples are "brace-at-the-end" style |
| 04:54.30 | trepan | ditto for ISO/IEC 14882/2003(E) <<C++>>, and ISO/IEC 23270:2006(E) <<C#>> |
| 05:04.48 | JeffM | I was just reading from wikipedia ;) |
| 05:06.04 | trepan | well now you know :) (what wiki page?) |
| 05:07.09 | JeffM | one on indenting |
| 05:07.22 | JeffM | it mentioned iso c under the allman section |
| 05:08.38 | trepan | cool, I see what you're talking about |
| 05:08.48 | trepan | I guess they upgraded for c99 ;) |
| 05:09.32 | JeffM | not that it will convince anyone to change |
| 05:09.42 | JeffM | there will be no one method that will make everyone happy |
| 05:10.03 | CIA-114 | BZFlag: 03JeffM2501 * r21513 10/trunk/bzflag/src/bzflag/ (HUDRenderer.cxx bzflag.cxx playing.cxx): warnings |
| 05:12.19 | trepan | hm, looking at a working draft for the original ANSI C spec, and it has braces at the end of lines too; i wonder if wikipedia has it wrong |
| 05:12.37 | JeffM | it's a consipriacy |
| 05:12.58 | trepan | damn kids with the purple bikesheds |
| 05:13.20 | JeffM | silly old people with terminals ;) |
| 05:13.36 | BulletCatcher | Since we're voting ;-), I prefer "gnu" style, which is like BZFlag's but with closing braces indented further. |
| 05:13.37 | BulletCatcher | There is no pleasing everyone. |
| 05:13.55 | JeffM | BulletCatcher, that's what allman does |
| 05:14.01 | JeffM | oh indented |
| 05:14.06 | JeffM | sorry |
| 05:15.00 | trepan | i thought that gnu style was like Allman, but with the braces shifed midway to the next indent level |
| 05:15.05 | meem1029 | Hmm, I prefer just using starting braces on the same line as class/function/if/else/etc. definition but closing ones on their own. I'm not really sure why I like that though. |
| 05:15.07 | JeffM | no, gnu is one one line |
| 05:15.13 | JeffM | acording to wikipedia |
| 05:15.32 | JeffM | meem1029, cus that's the way you learned |
| 05:16.09 | trepan | http://en.wikipedia.org/wiki/Indent_style -- Like the Allman and Whitesmiths styles, GNU style puts braces on a line by themselves, indented by 2 spaces, except when opening a function definition, where they are not indented.[7] In either case, the contained code is indented by 2 spaces from the braces. |
| 05:16.20 | JeffM | yeah it just sends it in more |
| 05:16.36 | meem1029 | Yep, pretty much. That would be BSD by looking at wikipedia. |
| 05:16.37 | JeffM | seems a little incosnistent, why make blocks in functions different then functions |
| 05:16.47 | JeffM | but hey at least they are on a line |
| 05:16.57 | BulletCatcher | Then that's the wrong gnu format. Let me find the name for my preference. |
| 05:17.24 | JeffM | you may be thinking of bsd knf |
| 05:17.36 | trepan | i tend to put the { on the same line as the function declaration, which is a little more consistent than the BSD style |
| 05:18.31 | JeffM | man curl is so messed up, if(filesize < -data->state.resume_from) |
| 05:18.43 | JeffM | resume_form is an unsigned type |
| 05:18.48 | BulletCatcher | "Banner" style is the name on that wikipedia page for my preference. |
| 05:19.04 | JeffM | ewwww |
| 05:19.07 | BulletCatcher | :-p |
| 05:19.14 | trepan | BulletCatcher: that's just weird :) |
| 05:19.43 | BulletCatcher | It isn't so weird that it didn't make it on the page. :-) |
| 05:19.47 | trepan | do a lot of python coding? |
| 05:19.53 | BulletCatcher | None at all. |
| 05:20.29 | CIA-114 | BZFlag: 03JeffM2501 * r21514 10/trunk/bzflag/MSVC/build/ (common.vcproj platform.vcproj): warnings |
| 05:20.38 | meem1029 | Hmm, I do my html tables like they have in the example (or did 3 years ago when I last did html), but always have my closing braces back the way normal people do. |
| 05:22.14 | BulletCatcher | When I write HTML by hand I tend not to indent at all. |
| 05:22.52 | BulletCatcher | That's gonna change when I start doing 4.01 strict HTML, though. |
| 05:23.00 | trepan | while we're painting, I abhor 'if(expr)' without the space; and prefer to use {} for every control statement, even the one-liners and switch cases |
| 05:23.04 | BulletCatcher | Too many divs. |
| 05:23.24 | JeffM | heh, I delete your one liners they piss me off |
| 05:23.57 | trepan | consistentcy, defenestrated :) |
| 05:24.02 | BulletCatcher | lol |
| 05:24.15 | JeffM | devinfo allows me to delete them ;) |
| 05:24.37 | trepan | devinfo wouldn't let you through the door ;) |
| 05:24.44 | JeffM | true |
| 05:25.03 | JeffM | but that's my story and I'm sticking to it |
| 05:25.08 | meem1029 | Ya, I generally use braces no matter how many lines the expression is. Depending on how lazy I'm feeling. |
| 05:25.09 | trepan | heh :) |
| 05:26.02 | BulletCatcher | My code tends to be compact. |
| 05:26.04 | BulletCatcher | I have been known to use expressions like "JanFebMarAprMayJunJulAugSepOctNovDec"[month * 3] . |
| 05:27.00 | BulletCatcher | But that was when I was young and less disciplined. |
| 05:28.54 | meem1029 | That reminds me of the alias I wrote to have a run script for a mud I play. The code for it is completely hacked together and terrible, but it works for a very limited subset of what I would ideally have it do that I decided I can make do with. |
| 05:29.38 | BulletCatcher | That is known as "good enough" code. |
| 05:29.55 | trepan | looks like -publickey never made it into the 2.99.x man page |
| 05:30.15 | trepan | oops, it did |
| 05:32.48 | trepan | (as -publicukey, as distant cousin to karaoke) |
| 05:32.48 | BulletCatcher | I am making slow progress on the build system, by the way. |
| 05:32.50 | BulletCatcher | At this point I am just trying to be able to use the packages under src/other. |
| 05:33.20 | JeffM | it seems like a very tedious task |
| 05:34.26 | CIA-114 | BZFlag: 03trepan * r21515 10/trunk/bzflag/man/bzfs.6.in: * added '-publickey' to the man page |
| 05:34.59 | BulletCatcher | The main problem is that I haven't internalized the autotools details as well as I'd like, but I suspect I'll be much better by the time I am done. |
| 05:35.39 | trepan | wonders if it would be faster to switch to cmake now, at least the winny world could be tied in |
| 05:36.43 | trepan | and maybe, just maybe, we could have the app. version defined in ONE place :) |
| 05:37.00 | BulletCatcher | Can CMake and autotools/make coexist? |
| 05:37.22 | ruskie | sure |
| 05:37.27 | trepan | sure, they both generate the makefiles |
| 05:37.41 | JeffM | yeah cmake can work, it's a huge pain on windows |
| 05:37.55 | JeffM | probably just use it and check in the projects anyway |
| 05:37.59 | JeffM | but I ain't doin it :) |
| 05:38.02 | BulletCatcher | Then there should be nothing to stop a motivated developer from adding it. |
| 05:38.31 | JeffM | it ususaly is used in place of make/autotolls |
| 05:38.43 | JeffM | and it place of vc projects |
| 05:38.52 | JeffM | cus it outputs makefiles and proejcts |
| 05:39.22 | JeffM | so if you had a functional makefile system and working projects, I don't see any reason to change to it |
| 05:39.33 | JeffM | it would only make sense if you were going to remove the other 2 |
| 05:40.03 | BulletCatcher | My idea would be to add CMake, proving that it works, before removing the others. |
| 05:40.19 | JeffM | that's a big project |
| 05:40.27 | BulletCatcher | So maybe not for 2.3, then. |
| 05:40.30 | JeffM | specialy to get it to make vc projects like I have set up |
| 05:41.26 | JeffM | I have never seen a cmake system work out of the box on windows |
| 05:41.42 | JeffM | that may be because people have not set it up right, but that does not give me much confidence on it |
| 05:42.41 | BulletCatcher | As in any language, there are more ways to do it wrong than there are to do it right. :-) |
| 05:42.42 | JeffM | the only reason that make changes were on the list for 2.3 was because they were already in 2.99 and proven to work |
| 05:44.49 | BulletCatcher | Right now I am trying to backport 2.99's configure.ac incrementally, but if I don't make progress soon I may just copy the 2.99 version and try fixing that. |
| 05:48.38 | meem1029 | I just looked through DEVINFO and found that the version number documentation is still using the old 3 number system. |
| 05:51.16 | JeffM | that should be fixed |
| 06:06.01 | meem1029 | Hmm, also looking around the files (I was looking to see if I would be able to fix it and looking for more info about the new versioning system) it appears that in /src/date/buildDate.cxx the new version numbering is not implemented. |
| 06:10.56 | CIA-114 | BZFlag: 03Meem1029 07http://my.bzflag.org * r7657 10/w/BZFlag_2.3: |
| 06:12.51 | BulletCatcher | I thought JeffM had already documented the new versioning system somewhere, but now I can't find it. |
| 06:13.43 | meem1029 | I know for sure that those 2 places do not have it documented/implemented, so I added it there. I would try to tackle it myself, but I don't understand enough about the code yet. |
| 06:14.42 | BulletCatcher | This particular item is mostly documentation and very little code. |
| 06:15.28 | BulletCatcher | And I'm still not sure what the 4th number is supposed to be for. |
| 06:16.19 | *** join/#bzflag Marzipan- (~Marzipan@p5B2266C9.dip.t-dialin.net) |
| 06:16.54 | meem1029 | Ya, primarily documentation, but I noticed at least that one file that appears to make use of it. I think it would be best if the code agrees with the documentation. And I don't remember what the reasons were for moving to a 4 number system. |
| 06:17.43 | *** join/#bzflag Marzipan (~Marzipan@bzflag/player/Marzipan) |
| 06:18.28 | BulletCatcher | Found it, at http://my.bzflag.org/w/Versions |
| 06:18.33 | meem1029 | Actually, I think it was made so that minor number could be the number that broke compatibility. I'm not 100% sure why we need 2 more numbers though. |
| 06:19.17 | meem1029 | Aha! That's it. |
| 06:21.02 | BulletCatcher | I still don't think that BZFlag needs the fourth number, but I'll have to argue that with JeffM. |
| 06:21.23 | meem1029 | shrugs |
| 06:21.30 | meem1029 | I'm not going to get into that. |
| 06:21.39 | BulletCatcher | You are smarter than I am. :-) |
| 06:31.50 | *** join/#bzflag bjornredtail (~amusselm@66-169-234-32.dhcp.snlo.ca.charter.com) |
| 06:36.15 | CIA-114 | BZFlag: 03trepan * r21516 10/trunk/bzflag/src/bzfs/ (ListServerConnection.cxx bzfs.cxx bzfs.h commands.cxx): * backported the '/owner' command (part of the -publickey feature set) |
| 06:56.51 | BulletCatcher | generates his public key at http://my.bzflag.org/listkeys/index.php |
| 07:08.45 | BulletCatcher | Well, the key works because my server shows up in the list, but the client just says "Error connecting to server." |
| 07:09.11 | BulletCatcher | Nothing in the server logs, not even a connection from global. |
| 07:16.29 | Constitution | I know this was like a week ago, but the ID flag has been broken in 2.0 for a little while at least... not sure what triggers it, but I've had the same issue with registering all flags as PZ |
| 07:17.34 | Constitution | alright nevermind, that was already known |
| 07:20.44 | BulletCatcher | Someone pointed out that it only affects zoneflag instances. |
| 07:22.34 | BulletCatcher | Using +f ID{2} in the options section apparently works correctly. |
| 07:30.03 | Constitution | ~mrapple++ |
| 07:32.38 | Constitution | also (forgive me again if this was discussed) I think brlcad fixed the list server IP caching in branch also |
| 07:53.06 | *** join/#bzflag Chestal (thilo@141.99.42.43) |
| 07:53.08 | *** mode/#bzflag [+o Chestal] by ChanServ |
| 10:38.07 | *** join/#bzflag Gilly (~btw@83.146.211.140) |
| 10:38.07 | *** join/#bzflag Gilly (~btw@about/essy/ilkimys/gilly) |
| 10:52.04 | *** join/#bzflag AAA_awright_ (~a3@ip24-251-157-63.ph.ph.cox.net) |
| 10:52.48 | *** join/#bzflag QBeep (~quantumbe@cpe-76-186-97-233.tx.res.rr.com) |
| 11:01.20 | *** join/#bzflag brad (~brad@stupid.h-o-s-t.name) |
| 11:45.59 | *** join/#bzflag brad (~brad@stupid.h-o-s-t.name) |
| 11:57.55 | *** join/#bzflag mdskpr_ (~mdskpr@108.25.108.121) |
| 12:43.01 | blast007 | Constitution: oh? hmm, let me check |
| 12:43.51 | blast007 | in my checkout of v2 it shows setDNSCachingTime(-1); |
| 12:44.16 | blast007 | so, nope :) |
| 12:48.20 | blast007 | meem1029: the reason for four numbers is so that we can fix packaging errors or tiny little flaws |
| 12:48.36 | blast007 | like how some projects include a patch level |
| 12:49.33 | blast007 | for instance, I think when we released 2.0.16, the Windows installer had a broken shortcut for the launching the client in a window since it was using the 2.99.x syntax (with -window taking the geometry settings) |
| 12:49.52 | blast007 | so we repackaged the installer under the same version number |
| 12:50.05 | blast007 | with a fourth number, we could have indicated it was slightly different |
| 12:50.25 | trepan | <BulletCatcher> Well, the key works because my server shows up in the list, but the client just says "Error connecting to server." -- twas broken before I arrived |
| 12:50.49 | trepan | hm, that a familiar statement, welcome to 2.99.x :) |
| 12:52.35 | blast007 | hmm, and maybe the bad shortcut was in 2.0.14 |
| 12:54.39 | mdskpr_ | blast007: if the bad shortcut was because of the -window geometry, then it was just broken by me a couple revisions ago |
| 12:54.41 | blast007 | and 1.10.8 was released just because the list server URL changed when the old server got hit by lightning... |
| 12:54.53 | short_circuit | How successful is a 2k8 EE build suppossed to be right now? I'm configured for debug win32 right now |
| 12:54.56 | blast007 | mdskpr_: no, I'm talking about our previous release |
| 12:55.04 | blast007 | 2.0.14 had a bad shortcut |
| 12:55.27 | blast007 | cuz we brought the NSIS installer from 2.99.x back into that code ;) |
| 12:56.13 | blast007 | (and since we brought the 2.99.x installer back into 2.3, it's all good - that one had the -window NNNxNNN already) |
| 13:31.20 | Thumper_ | the last successful connection to a server from linux was at r21500 for me |
| 13:33.06 | *** join/#bzflag brad (~brad@stupid.h-o-s-t.name) |
| 13:44.26 | blast007 | spldart: are you getting build errors? |
| 14:03.27 | *** join/#bzflag eTangenT (~lamer@c-98-245-170-238.hsd1.co.comcast.net) |
| 14:10.56 | *** join/#bzflag bier|tp (~bier@p57A66680.dip.t-dialin.net) |
| 14:15.23 | spldart | Yeah. 26 errors, 154 warnings. If that isn't suppossed to be the case then I need to know what I have configured incorrectly |
| 14:21.51 | blast007 | well I can't tell you that without knowing the errors |
| 14:21.57 | blast007 | pastebin them |
| 14:22.04 | blast007 | or at least some of them |
| 14:22.19 | blast007 | are there any about dinput.h not being found? |
| 14:22.59 | *** join/#bzflag Upsetter (~er@i59F6C1EC.versanet.de) |
| 14:30.52 | trepan | I just backported the 2.99.x dynamic colors, and got to thinking; if we put as much effort into fixing 2.99.x as I've seen recently trying to make 2.3 work, we'd already be done :) |
| 14:34.39 | BulletCatcher | I was hoping to not require map changes for 2.3. |
| 14:34.53 | spldart | http://bzflag.pastebin.ca/2062527 |
| 14:35.25 | trepan | BulletCatcher: it's a useful feature, a simplifies the syntax |
| 14:35.32 | trepan | *and simplifies |
| 14:36.02 | BulletCatcher | I think you are the only person who ever understood how to upgrade maps to the new syntax. |
| 14:36.48 | trepan | then noone's spent more than 30 seconds trying to figure out, it's trivial |
| 14:36.53 | BulletCatcher | It is not. |
| 14:37.12 | mdskpr_ | seems like i should write some doc |
| 14:37.16 | trepan | It is. |
| 14:37.45 | trepan | a sequence of color states with duration, much simpler than sinusoidal composition |
| 14:38.26 | mdskpr_ | so a list of colors that it fades between? |
| 14:39.55 | trepan | ramp 2 red \n ramp 4 blue \n -- spend 2 seconds ramping from red to blue, then 4 seconds ramping from blue to red |
| 14:43.32 | trepan | and the dynamic color BZDB tie-in allows you to create zone based gameplay, like the tic-tac-toe setup I had a while ago (but tic-tac-toe hacked at GameTime to get the color changes, a complicated setup) |
| 14:44.58 | BulletCatcher | Map makers will need clear instructions if their maps must be modified for 2.4. |
| 14:45.03 | BulletCatcher | Perhaps a wiki page, with examples, would satisfy that. |
| 14:45.05 | BulletCatcher | But it is unrealistic to expect others to understand this as easily as you do, trepan. |
| 14:45.53 | trepan | iirc, delusional already did a converted version of mofo without much ado |
| 14:47.10 | BulletCatcher | Did he take advantage of any of the new capabilities, or did he just change because he had no choice? |
| 14:47.25 | trepan | for 3.0, the latter |
| 14:51.03 | BulletCatcher | I see this kind of map change as more of an obstacle to 2.4 deployment than a benefit. |
| 14:52.38 | mdskpr_ | BulletCatcher: have you figured out a workaround for connecting to servers on linux? |
| 14:53.37 | BulletCatcher | No. JeffM's r21511 seems to have broken everything. |
| 14:54.00 | trepan | BulletCatcher: fair enough, I'm all done then :) |
| 14:54.22 | BulletCatcher | s/r21511/r21501/ |
| 14:58.49 | BulletCatcher | You're a man ahead of your time, trepan. |
| 14:58.53 | BulletCatcher | Thanks for the contributions you did make. |
| 14:58.54 | BulletCatcher | After 2.4 is out I think we will be receptive to more disruptive changes. |
| 14:59.47 | trepan | after 2.4, then what? start tacking 3.0 features back in? reboot the project? |
| 15:01.43 | BulletCatcher | We haven't planned that far ahead. |
| 15:28.01 | *** join/#bzflag fahadsadah (fahad@unaffiliated/fahadsadah) |
| 15:34.33 | meem1029 | Hmm, one of the recent updates seems to have broken all connectivity for me on Linux (before I was able to connect using ip addresses, just not the server name). |
| 15:40.48 | blast007 | meem1029: yeah, it was the HTTP-style header change |
| 15:40.56 | trepan | meem1029: <BulletCatcher> No. JeffM's r21501 seems to have broken everything. ? |
| 15:41.54 | blast007 | thankfully nobody plays on linux anyway |
| 15:41.56 | trepan | looks like the header change was in r21502 |
| 15:41.58 | blast007 | ducks |
| 15:42.01 | meem1029 | Ya, I remember one night (2 nights ago?) I was having problems with the domains while someone (BulletCatcher I think) was able to connect fine in Linux. |
| 15:42.08 | meem1029 | throws something at blast. |
| 15:42.21 | blast007 | I already ducked. So it missed. |
| 15:42.26 | trepan | linux, silly OS, power-user wannabes tweaking the bits and the bytes |
| 15:43.07 | meem1029 | Honestly, I just installed it to have as a backup in case windows decided to get a virus on me and then liked it so now I hardly ever boot into windows. |
| 15:43.16 | blast007 | :) |
| 15:43.36 | blast007 | I just have too many Windows games.. |
| 15:46.12 | trepan | switched to linux for the flexibility, but stayed for the pr0n |
| 16:07.44 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 16:07.44 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 16:10.33 | meem1029 | Hmm, this is confusing. On 2.0.16 I get a segfault when connecting to the purple panzer quickshot server. |
| 16:14.06 | blast007 | meem1029: with a client compiled by you? or from the distro's repo? |
| 16:14.53 | meem1029 | I'm pretty sure it's from Ubuntu repo. And this has never happened to me before (that's the map I play the most). Oh well, it's probably my computer telling me to do my homework. |
| 16:24.43 | delusional | You gotta break a few bzw's to make an omelette. |
| 16:24.56 | delusional | I say anything trepan can get in, get it in |
| 16:25.13 | delusional | wait, that doesnt sound righ |
| 16:25.14 | delusional | t |
| 16:25.54 | trepan | my pr0n comment got you wired for weird? :) |
| 16:26.29 | delusional | Break the maps if you have to... The first thing we did was make an improvement to the billboards that we wanted to do, but couldnt |
| 16:26.38 | delusional | nice fade from image to image |
| 16:28.21 | trepan | leaves it to BulletCatcher to decide, I have the diff saved |
| 16:29.10 | delusional | THis is exactly what i was talking abut earlier... there has to be things that the casual player can see with his own eyes |
| 16:32.14 | blast007 | exactly. we need a cow launcher flag. |
| 16:36.02 | meem1029 | Hmm, shoots a projectile along an arc (similar to how jumping goes, maybe the same as a jump) that explodes in a shockwave when it hits something? |
| 16:43.00 | blast007 | meem1029: the cow launcher flag is just a joke I bring up every so often :P |
| 16:43.10 | meem1029 | I know :P |
| 16:43.17 | blast007 | I can't recall who came up with the idea |
| 16:43.21 | blast007 | it was yeaaars ago |
| 16:43.32 | meem1029 | It would still be pretty awesome. |
| 17:04.54 | *** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501) |
| 17:04.54 | *** mode/#bzflag [+v JeffM] by ChanServ |
| 17:09.12 | blast007 | JeffM: linux isn't apparently cooperating with the HTTP-style header change for some reason |
| 17:11.06 | JeffM | there was non windows code that I did not change |
| 17:11.10 | JeffM | someone will have to debug it |
| 17:12.01 | JeffM | how does it fail? |
| 17:12.05 | JeffM | does something crash? |
| 17:12.17 | mdskpr_ | it doesnt let you connect to any server |
| 17:12.25 | JeffM | that isn't enough info |
| 17:12.44 | JeffM | does the server crash? does it time out? does it say connection failed? |
| 17:12.59 | mdskpr_ | It reports "Error: Connecting to Server" it times out i guess |
| 17:13.11 | JeffM | how long does it take to say that message? |
| 17:13.28 | JeffM | is it fast or does it take like 30 seconds? |
| 17:14.03 | mdskpr_ | about 10 seconds |
| 17:14.08 | mdskpr_ | after it gets the token |
| 17:14.48 | mdskpr_ | server list does not work either |
| 17:14.52 | JeffM | put a breakpoint on the linux server in processConnectedPeer |
| 17:15.04 | JeffM | see if the client sends the header |
| 17:15.31 | JeffM | see if it gets to the part that says // the dude has sent SOME data |
| 17:15.55 | JeffM | is it possible the issues are related to ares/curl? |
| 17:16.10 | JeffM | does the client ever get into the ServerLink class? |
| 17:16.38 | *** join/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 17:16.40 | mdskpr_ | idk |
| 17:16.48 | JeffM | these are things that need to be debuged |
| 17:17.33 | JeffM | the constructor for the SeverLink class on the client is what sends the header, make sure that it's getting there. |
| 17:17.49 | JeffM | if it's not getting there then it's ares that is probably not resolving the domain name |
| 17:17.57 | JeffM | and that's probably build system stuff |
| 17:18.12 | JeffM | if it gets there and sends the header, put breakpoints in the server to see if it gets it |
| 17:18.26 | mdskpr_ | its not ares, because IPs dont work |
| 17:18.40 | JeffM | it could be, we do reverse dns on stuff |
| 17:20.39 | JeffM | the server will do a reverse dns lookup on the connected player before it adds it as a player |
| 17:20.43 | JeffM | if that fails then we never add them |
| 17:20.46 | JeffM | and it times out |
| 17:20.52 | JeffM | so it could be failing on that side |
| 17:20.57 | JeffM | thse are things you can check |
| 17:21.23 | JeffM | once the client sends the header, the server calls MakePlayer and turns the nethandler from a peer into a player |
| 17:21.30 | JeffM | and that starts the reverse dns lookup |
| 17:22.13 | JeffM | when it finishes the server then gets notified in doStuffOnPlayer and sends the normal connection string back ( the server version ) |
| 17:24.45 | JeffM | I had issues when developing it of the client just sitting there because the server never added them as a player, and I had to put a lot of breakpoints to make sure that everything happened in order |
| 17:25.02 | JeffM | so you should be able to do the same thing on linux and see where it's failing |
| 17:25.12 | mdskpr_ | ok |
| 17:25.25 | JeffM | you can also put a breakpoint in removePlayer to see if the server is rejecting the connection |
| 17:25.32 | JeffM | the debuger is your best friend :) |
| 17:33.01 | mdskpr_ | it got to the part where the dude sent data |
| 17:34.11 | JeffM | ok then does it parse correctly and call makePlayer |
| 17:34.15 | JeffM | step on in |
| 17:36.12 | JeffM | if it gets in there ok, then see if it hits line 4648 iun bzfs.cxx, in doStuffOnPlayer |
| 17:36.16 | JeffM | the call to addPlayer |
| 17:36.22 | *** part/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 17:36.25 | JeffM | I bet it never gets there |
| 17:37.51 | mdskpr_ | nope |
| 17:38.05 | JeffM | nope as in it got there? or nope it did not get there? |
| 17:38.08 | mdskpr_ | dies before the while loop in processPeerConnected |
| 17:38.18 | JeffM | like crashes? |
| 17:38.24 | mdskpr_ | no, just stops |
| 17:38.59 | JeffM | where does it "stop"? |
| 17:39.10 | JeffM | what is in the tcp buffer? |
| 17:39.18 | mdskpr_ | line 4910, idk? |
| 17:39.29 | JeffM | did you step line by line and see what it did? |
| 17:40.04 | mdskpr_ | in every section, not line by line |
| 17:40.16 | JeffM | you need to go line by line and look at what the code is doing |
| 17:40.25 | JeffM | how much data did it get on the first read? |
| 17:40.31 | JeffM | was it the 8 character string |
| 17:40.38 | mdskpr_ | where do you get that info? |
| 17:40.39 | JeffM | cust 4910 should never be hit for regular players |
| 17:40.56 | JeffM | your debuger should let you see the variables in scope |
| 17:41.11 | JeffM | and have a watch section to let you see the values of specific variables |
| 17:41.16 | JeffM | I don't know how to use your debuger |
| 17:41.39 | mdskpr_ | nor me : / |
| 17:41.40 | JeffM | readSize should be 8 |
| 17:41.49 | JeffM | this is a tool you need to know for development |
| 17:43.24 | JeffM | if you can't see code as it's running you can't fix problems in it |
| 17:43.36 | JeffM | that or you end up using a bunch of printfs |
| 17:44.01 | mdskpr_ | yeah that is waht i was doing just now, a bunch of logDebugMessage() |
| 17:44.19 | JeffM | not the ways the pros do it ;) |
| 17:44.41 | JeffM | I just press f9 to set a breakpoint then use F10 and F11 to step over or in |
| 18:05.37 | *** join/#bzflag temporalD (~a_temp_di@bzflag/serverop/TemporalDistraction) |
| 18:06.03 | blast007 | wonder what options there is for a graphical interface to GDB. never bothered looking. |
| 18:06.23 | JeffM | there arem any IIRC |
| 18:08.15 | blast007 | KDbg, though that's for KDE.. wonder if there's a Gnome equivilent |
| 18:08.34 | JeffM | you don't have to insdtall full kde to run kde apps |
| 18:08.59 | blast007 | still |
| 18:09.00 | blast007 | there's this too: http://www.gnu.org/software/ddd/ |
| 18:09.44 | JeffM | Nemiver and dd |
| 18:09.45 | JeffM | yeah |
| 18:10.40 | *** part/#bzflag Upsetter (~er@i59F6C1EC.versanet.de) |
| 18:10.56 | blast007 | Nemiver looks simple enough to use |
| 18:11.27 | mdskpr_ | hmmm, readSize is 0 |
| 18:11.41 | JeffM | that's the problem |
| 18:11.42 | mdskpr_ | at 4878\ |
| 18:13.03 | JeffM | find out why |
| 18:13.32 | blast007 | JeffM: was bzadmin updated to use the new connect method? |
| 18:13.41 | JeffM | I thoughyt it used serverlink |
| 18:14.03 | blast007 | possible, just wasn't sure |
| 18:14.21 | JeffM | I assumed it did |
| 18:14.28 | JeffM | cus I didn't see anthing else that did it |
| 18:14.39 | JeffM | and if it dosn't it should ;) |
| 18:15.06 | blast007 | I didn't have a graphical client here, otherwise I'd just test with that |
| 18:16.25 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 18:16.25 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 18:29.03 | blast007 | JeffM: for me with bzadmin, 'netHandler->receive(strlen(BZ_CONNECT_HEADER))' is returning ReadDiscon |
| 18:29.08 | blast007 | at least now |
| 18:29.21 | JeffM | suck |
| 18:29.48 | JeffM | put a breakpoint in serverlink and see if it sends it |
| 18:30.16 | blast007 | ok |
| 18:30.32 | JeffM | the change was failry simple conceptulay |
| 18:30.49 | JeffM | what used to happen was the client wold connect and wait for the server to send out it's version |
| 18:30.51 | blast007 | oh, and it seems like bzadmin has it's own ServerLink stuff |
| 18:31.11 | blast007 | so maybe this test isn't going to go anywhere ;) |
| 18:31.17 | JeffM | now right after connect the client sends out the connect header. The server waits untill it has the connect header before sending the version |
| 18:31.20 | JeffM | after that its' all the same |
| 18:31.29 | JeffM | oh that sucks |
| 18:31.35 | JeffM | then yes that has to be updated |
| 18:31.39 | JeffM | I can look at that in a bit |
| 18:32.39 | JeffM | well maybe more then a bit, forgot I left my API totaly busted :) |
| 18:32.41 | blast007 | though it seems they're identical, unless I'm doing it wrong. |
| 18:33.10 | JeffM | did it just copy it? |
| 18:33.15 | JeffM | if so make it use the one in the client |
| 18:33.49 | JeffM | gotta go to lunch |
| 18:39.30 | blast007 | it doesn't seem like bzadmin is sending anything to the port - it's just sitting there waiting |
| 18:40.13 | blast007 | mdskpr_: can you try connecting to bzexcess.com (173.45.72.146) on port 5000 with a 2.3 client? |
| 18:40.20 | blast007 | one with the HTTP header thing |
| 18:40.23 | blast007 | the graphical client |
| 18:42.33 | blast007 | (or anyone with an up-to-date graphical client) |
| 18:42.51 | blast007 | I just have a copy of netcat sitting there's listening for a connection |
| 18:45.45 | mdskpr_ | ok done |
| 18:46.26 | blast007 | hmm, yeah, nothing. |
| 18:46.51 | blast007 | anyone with a Windows client of the latest 2.3? |
| 18:47.38 | mdskpr_ | you sure its alive? a 2.0.16 client would have at least said incompatibly version |
| 18:47.48 | blast007 | mdskpr_: it wasn't a server |
| 18:47.55 | mdskpr_ | oh |
| 18:47.58 | blast007 | it was just netcat listening on 5000 |
| 18:48.16 | mdskpr_ | did you get anything when i tried it with 2.0.16? |
| 18:48.16 | blast007 | if 2.3 was working right, it should have spit out a BZFLAG on my end |
| 18:48.26 | blast007 | 2.0.16 wouldn't |
| 18:48.41 | blast007 | 2.0.x waits for the "BZFS0026" to get sent to it before it sends anything |
| 18:48.46 | mdskpr_ | oh |
| 18:49.11 | blast007 | plus, netcat (at least the way I ran it) only stays there for one connection |
| 18:49.25 | blast007 | so did you connect with a 2.3 one initially? |
| 18:49.32 | mdskpr_ | yes |
| 18:49.37 | blast007 | okay |
| 19:43.43 | JeffM | there is only one serverlink file and it looks like bzadmin uses it, at least on windows |
| 19:45.13 | blast007 | JeffM: a graphical client that someone joined didn't send the BZFLAG\r\n\r\n to a netcat instance, nor did my local bzadmin |
| 19:45.44 | JeffM | put a breakpoint in serverlink |
| 19:45.47 | JeffM | and see what happens |
| 19:46.30 | JeffM | line 212 |
| 19:46.36 | JeffM | <PROTECTED> |
| 19:46.55 | JeffM | ahh I see the issue |
| 19:48.05 | JeffM | svn up |
| 19:48.08 | CIA-114 | BZFlag: 03JeffM2501 * r21517 10/trunk/bzflag/src/bzflag/ServerLink.cxx: linux has an EXTRA check for the server version, put that after the connect header |
| 19:49.34 | JeffM | if that dosn't work then try just commenting out lines 258 to 268 |
| 19:49.39 | JeffM | I'm not sure it needs to check again |
| 19:49.40 | McSpider | works fine on Mac OSX now |
| 19:50.15 | JeffM | can you comment out those lines and see if it still works? |
| 19:50.21 | JeffM | does not like dupe code |
| 19:52.29 | McSpider | yep, still works |
| 19:55.29 | JeffM | k, once someone tells me that it works on linux too I'll remove the code |
| 19:56.35 | McSpider | When I leave the server I get: |
| 19:56.36 | McSpider | Program received signal SIGPIPE, Broken pipe. |
| 19:56.37 | blast007 | it worked for me, but I'm doing a full rebuild with clean sources just to make sure it wasn't something screwy with that duplicate ServerLink.cxx |
| 19:56.37 | McSpider | 0x00007fff800c7e3e in sendto () |
| 19:57.00 | JeffM | debug and find out why |
| 19:57.09 | JeffM | client cleanup should work the same |
| 19:57.21 | JeffM | blast007, ok |
| 19:57.38 | blast007 | also, I'm only testing with bzadmin, not the full client |
| 19:57.44 | JeffM | goes back to breaking the heck out of the API |
| 19:57.49 | blast007 | but I can test with the full in around 2 hours |
| 19:57.51 | JeffM | my gut tells me I can remove that code |
| 19:57.55 | JeffM | I just want some more testing |
| 19:58.27 | JeffM | interplugin communication is going to be super simple now ;) |
| 19:58.45 | McSpider | only get that eerror if im playing with -solo bots |
| 19:58.59 | blast007 | yeah, it works without that code, JeffM |
| 19:59.11 | JeffM | blast007, can you think of any reason why anyone would NEED to load the same plugin more then once? |
| 20:00.01 | blast007 | when they're lame like me and write a plugin that accepts a bzid instead of a config file and thus can't pass in multiple players? ;) |
| 20:00.15 | blast007 | other then that, no, can't think of a reason |
| 20:00.18 | JeffM | do you think that's viable? |
| 20:00.28 | JeffM | cus I'm giving plugins a registerd "name" |
| 20:00.39 | JeffM | and I can make it so they are unique |
| 20:00.49 | JeffM | or I can make uniqueness optional |
| 20:00.55 | JeffM | cus many plugins need to be unique |
| 20:00.58 | JeffM | like SW death |
| 20:01.13 | blast007 | yeah, just allow each plugin once |
| 20:01.16 | JeffM | ok |
| 20:01.30 | blast007 | if there's a case that someone wants to have a plugin loaded twice, then that's a sign that the plugin is flawed, not our system |
| 20:01.43 | JeffM | there will also be a "find plugin" API call so you can get the class for another plugin and call stuff in it |
| 20:01.48 | JeffM | this will make HTTP way simpler |
| 20:01.50 | blast007 | awesome |
| 20:02.19 | JeffM | and we can have each plugin set it's minium idle wait time, and I can make the bz loop do the shortest |
| 20:02.31 | JeffM | bassicly we have a class per plugin now |
| 20:03.18 | JeffM | and the API version is now the minium version needed not the maxium |
| 20:03.33 | JeffM | so we don't load new plugins but we'll load old ones |
| 20:03.41 | JeffM | well old 2.4 ones :) |
| 20:08.00 | *** join/#bzflag Erroneous (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 20:08.01 | *** mode/#bzflag [+v Erroneous] by ChanServ |
| 20:52.10 | CIA-114 | BZFlag: 03JeffM2501 * r21518 10/trunk/bzflag/src/bzflag/ServerLink.cxx: we use common code to get the server version, don't need unix specific stuff. |
| 21:04.37 | trepan | JeffM: with r21518, can you start a local server and connect to it using the windows build? |
| 21:04.51 | JeffM | can't test for another 4 hours |
| 21:04.58 | JeffM | so dunno |
| 21:05.27 | JeffM | I was curious if linux needed that code or not |
| 21:05.44 | trepan | looks like the change allows me to connect, cool |
| 21:05.58 | JeffM | other had said it did as well |
| 21:06.04 | meem1029 | Well, I just downloaded the new patch and it didn't let me connect. Currently recompiling after a make clean. |
| 21:06.16 | JeffM | the code I changed was only in a non windows #def |
| 21:06.21 | JeffM | so I belive windows still works |
| 21:06.27 | JeffM | but I have not been able to test it |
| 21:08.41 | meem1029 | Yep, I'm having connection problems with r21518 |
| 21:09.02 | JeffM | roll it back one rev and see if that helps |
| 21:09.06 | JeffM | there were 2 commits I made |
| 21:09.23 | JeffM | the first one got more people to connect, it is the second one I am unsure of |
| 21:09.29 | mdskpr_ | JeffM: the length of data bzflag is sending is 10,*NetHandler.cxx:553..it is sending too much data |
| 21:09.39 | JeffM | why is it doing that? |
| 21:09.53 | mdskpr_ | dunno yet |
| 21:10.02 | JeffM | oh wait, it shuld be 10 |
| 21:10.08 | JeffM | BZFLAG\r\n\r\n |
| 21:10.22 | JeffM | we use strlen in the server |
| 21:10.24 | mdskpr_ | hmmm |
| 21:10.25 | JeffM | so good, it should be 10 |
| 21:10.30 | JeffM | does it call MakePlayer? |
| 21:10.48 | mdskpr_ | no |
| 21:10.54 | JeffM | why not? |
| 21:11.01 | mdskpr_ | RxStatus is ReadPart |
| 21:11.21 | JeffM | it sent 10 and read 10, that sould be ReadALL |
| 21:11.31 | JeffM | what's keeping it open? |
| 21:11.51 | mdskpr_ | yet, there is an if statement...if length<= tcplen(which is 0) then it is readAll |
| 21:12.01 | mdskpr_ | 10<=0 not true.... |
| 21:12.11 | mdskpr_ | who tells what tcplen is? |
| 21:12.39 | JeffM | nethanlder |
| 21:12.53 | JeffM | unsigned int readSize = netHandler->getTcpReadSize(); |
| 21:12.55 | JeffM | that returns 10? |
| 21:12.59 | mdskpr_ | no 0 |
| 21:13.09 | mdskpr_ | because tcplen is 0 |
| 21:13.19 | mdskpr_ | but the data received is 10...i think i have a workaround |
| 21:13.22 | JeffM | hmm |
| 21:13.46 | JeffM | so the getTcpBuffer is 10? |
| 21:14.34 | JeffM | does it need a little loop to get to readall before calling the tcp size? |
| 21:15.04 | mdskpr_ | idk |
| 21:15.09 | JeffM | if it returns part it just means it read less then I asked it to read |
| 21:15.12 | mdskpr_ | but this returns 0: RxStatus e = netHandler->receive(strlen(BZ_CONNECT_HEADER)); |
| 21:15.36 | JeffM | did the client send the data? |
| 21:15.47 | mdskpr_ | yes |
| 21:16.01 | JeffM | 0 is readall |
| 21:16.24 | JeffM | did you update with my code I commited? |
| 21:16.36 | JeffM | and did it merge with all your printf's? |
| 21:16.38 | mdskpr_ | yeah |
| 21:16.42 | JeffM | you sure |
| 21:16.45 | JeffM | others are connecting better |
| 21:16.47 | mdskpr_ | i got rid of the printfs |
| 21:16.50 | mdskpr_ | im using gdb now |
| 21:16.56 | JeffM | but I don't know |
| 21:17.24 | JeffM | as far as I understand the RxStatus should be all or part if some data was read |
| 21:17.36 | mdskpr_ | it was part |
| 21:17.40 | meem1029 | What is the command to do a completely clean make of it (--development-clean or something similar, but that doesn't work)? |
| 21:17.42 | JeffM | 0 is not a part |
| 21:17.48 | mdskpr_ | make clean |
| 21:17.54 | JeffM | 0 is nothing |
| 21:18.11 | JeffM | all or part should return something between 1 and 10 |
| 21:18.36 | trepan | 'make distclean' with also clean out some of the build config files |
| 21:18.39 | JeffM | part would be that I asked for 10 but got some data but not enough for 10 |
| 21:18.51 | mdskpr_ | ok |
| 21:19.33 | JeffM | and part should be handled, it builds up a buffer with all the data untill it gets to 10 |
| 21:19.36 | JeffM | then checks the buffer |
| 21:19.39 | trepan | 'svn-clean' is also useful |
| 21:19.44 | JeffM | unless something in the API grabs it first |
| 21:20.37 | trepan | are you accounting for the fact that MsgGameTime packets start being sent immediately (or used to be) |
| 21:20.47 | JeffM | they are not sent yet |
| 21:21.13 | JeffM | I put this code on the client before it waits for the server version |
| 21:21.26 | mdskpr_ | ah, i got it to return ReadAll now |
| 21:21.36 | JeffM | and the game time is sent from the server in makePlayer, that's called AFTER the conenct buffer |
| 21:21.46 | JeffM | so I belive I am accounting for that |
| 21:21.59 | trepan | MsgLagState? |
| 21:22.06 | JeffM | no messages should be sent |
| 21:22.12 | JeffM | from the server or client |
| 21:22.22 | trepan | when I 'nc' to a server, I'm seeing 'lslsls...' |
| 21:22.22 | JeffM | on the server they arn't even in the player list yet |
| 21:22.47 | JeffM | I keep a seperate list of net handlers for players still connecting |
| 21:22.55 | JeffM | unless nethandler sends those out itself |
| 21:23.01 | *** join/#bzflag KTL (~KTL@213.219.156.22.adsl.dyn.edpnet.net) |
| 21:23.35 | JeffM | the nethandler pointer is transfered into the player class, at that point the old game logic takes over and does whatever it used to do. |
| 21:24.53 | JeffM | and on the client it's doing a blocking wait untill the server version comes back |
| 21:25.01 | JeffM | so I don't think the client is sending anything |
| 21:30.09 | trepan | heh, those 'ls' were because my 'bzfs' executable is still 2.99.x, sorry |
| 21:30.35 | JeffM | heh |
| 21:31.07 | trepan | MsgLagState is defined in 2.3's Protocol.h, but it isn't even used |
| 21:31.09 | JeffM | yeah I did a fair ammount of debuging of the general process on windows so I feel it's good. it was just that code in the non windows #def that I could not test and kinda glossed over. |
| 21:31.16 | JeffM | fun |
| 21:40.00 | meem1029 | Hmm, very interesting. I'm running the last patch (21517) and am able to connect to a server running locally using 127.0.0.1 but not using localhost |
| 21:40.02 | *** join/#bzflag ahs3- (~quassel@adsl-065-005-193-158.sip.rdu.bellsouth.net) |
| 21:40.46 | *** join/#bzflag ahs3- (~ahs3@adsl-065-005-193-158.sip.rdu.bellsouth.net) |
| 21:42.32 | mdskpr_ | indeed |
| 21:43.54 | JeffM | that's ares |
| 21:44.01 | JeffM | and part of the build system problems |
| 21:44.05 | JeffM | install windows, you'll be fine ;) |
| 21:44.27 | mdskpr_ | can now resume more backports |
| 21:44.39 | JeffM | did you see the ones I assigned to you? |
| 21:45.14 | mdskpr_ | no |
| 21:45.26 | JeffM | I assigned 2 patches to you to apply on SF |
| 21:45.35 | JeffM | the ones from mcspider |
| 21:45.41 | mdskpr_ | ah, i never check my email |
| 21:47.36 | mdskpr_ | i already did the no-tks, but didnt commit it cuz i couldnt test it |
| 21:48.02 | McSpider | My patches are tested btw |
| 21:48.16 | JeffM | then there ya go ;) |
| 21:48.18 | mdskpr_ | well, you have some funny stuff with the BZ_Rev numbers |
| 21:48.29 | mdskpr_ | one file says 5, the other 4 |
| 21:48.47 | McSpider | hmm, odd |
| 21:49.03 | mdskpr_ | you dont have to update it |
| 21:49.15 | JeffM | yeah that part you will have to do manualy |
| 21:49.32 | JeffM | for the no tk, yes you do |
| 21:49.35 | JeffM | since it changes death |
| 21:49.40 | JeffM | bump the proto too |
| 21:49.46 | mdskpr_ | yeah i was gonna |
| 21:50.39 | meem1029 | I was about to offer to help test no tk, but then I realized that I wouldn't be much use since I can't even get my client to connect to a server properly. |
| 21:51.20 | McSpider | I also have a patch for OpenFFA, but I need to test it some more. |
| 22:04.28 | CIA-114 | BZFlag: 03mudskipper * r21519 10/trunk/bzflag/ (8 files in 6 dirs): Accepted patch-3303071 from McSpider adding -noTeamKills option to bzfs. |
| 22:04.53 | JeffM | did you add McSpider to the authors file? |
| 22:05.02 | mdskpr_ | ah right |
| 22:05.07 | JeffM | ;) |
| 22:08.08 | CIA-114 | BZFlag: 03mudskipper * r21520 10/trunk/bzflag/AUTHORS: Added McSpider's name to the AUTHORS file. |
| 22:12.09 | *** join/#bzflag a_meteorite (~bzmeteori@unaffiliated/ameteorite/x-000000001) |
| 22:13.22 | blast007 | JeffM: connecting to "localhost" doesn't work on windows either (but 127.0.0.1 does) ;) |
| 22:13.39 | JeffM | I was using localhost in my testing at home |
| 22:13.45 | JeffM | I can test it there when I get home |
| 22:13.51 | JeffM | in 3 hours |
| 22:14.09 | blast007 | windows 7? |
| 22:14.15 | JeffM | yeah |
| 22:14.18 | CIA-114 | BZFlag: 03mudskipper * r21521 10/trunk/bzflag/ (4 files in 2 dirs): Accepted patch - 3303071 from McSpider, removing client capability to turn off fog |
| 22:14.44 | blast007 | guess I'll try again |
| 22:15.25 | blast007 | I did notice a note in the Windows 7 hosts file that "localhost name resolution is handled within DNS itself." |
| 22:16.04 | mrapple | will the irational team flag patch make it in? |
| 22:16.04 | blast007 | I'll build again. maybe it was just a problem before the ares tweaks. |
| 22:16.44 | mdskpr_ | irrational? |
| 22:16.58 | trepan | pi! |
| 22:17.07 | JeffM | everyone loves pi |
| 22:24.39 | PlasticTank | Aww, I hate fog -_- |
| 22:25.38 | JeffM | then don't play on servers with fog ;) |
| 22:30.42 | trepan | or compile your own client -- we'll never know :) |
| 22:32.57 | PlasticTank | \ |
| 22:33.02 | PlasticTank | opps |
| 22:46.43 | mdskpr_ | JeffM: do you mind if -requireUDP is removed? |
| 22:47.35 | JeffM | hell no |
| 22:47.38 | JeffM | do it |
| 22:52.00 | CIA-114 | BZFlag: 03mudskipper * r21522 10/trunk/bzflag/ (7 files in 4 dirs): removed -requireUDP, it is always required now. |
| 22:57.58 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 23:02.23 | mdskpr_ | are there any functions to change a player's team ingame? |
| 23:04.26 | trepan | in 2.99.x, yes |
| 23:45.07 | JeffM | man the old API was screwed up |
| 23:45.17 | JeffM | you HAD to recompile every time the API version changed |
| 23:57.38 | meem1029 | sighs |
| 23:58.14 | meem1029 | My 2.0.x client is still segfaulting on quickshot. Guess I'll have to go dl the source and recompile. |