| 00:04.39 | JeffM | curl won't link then |
| 00:04.54 | JeffM | hmm wonder if it just needs the dependency |
| 00:06.00 | JeffM | we don't use curl for server connections |
| 00:06.10 | JeffM | in fact curl is working cus the list server works |
| 00:06.18 | JeffM | so it has to be with the client |
| 00:08.25 | blast007 | there are additional libs that 2.99.x's bzflag project links with: Advapi32.lib user32.lib Gdi32.lib shell32.lib |
| 00:08.48 | blast007 | but I wouldn't think that should matter if it still builds the .exe, right? |
| 00:09.02 | JeffM | yeah |
| 00:09.15 | JeffM | cus we sure as heck don't use GDI |
| 00:23.10 | JeffM | it's because aresFailed is true |
| 00:23.22 | JeffM | bzflag.exe calls ares to resolve the name |
| 00:23.30 | blast007 | yeah |
| 00:24.32 | JeffM | I think we have to call something for windows first |
| 00:25.33 | JeffM | ares_win32_init |
| 00:25.39 | JeffM | where is that called in 2.99 |
| 00:26.22 | blast007 | the src/net/AresHandler.cxx has a bit more in the constructor |
| 00:26.32 | blast007 | specifically a section labeled "start up our resolver" |
| 00:26.40 | JeffM | we probalby need that |
| 00:26.40 | blast007 | well, wait |
| 00:26.56 | blast007 | 2.3 has that same comment, but with less code |
| 00:26.57 | DTRemenak | there were a LOT of changes for ares/curl in 2.99 |
| 00:27.10 | DTRemenak | you should probably just grab the whole set of files |
| 00:27.16 | JeffM | hopefuly we can just move over the entire handler |
| 00:27.16 | JeffM | yeah |
| 00:29.38 | JeffM | yeah there is a HAVE_ARES_LIBRARY_INIT |
| 00:30.46 | JeffM | if the enire handler won't move over I can probably fix it |
| 00:31.40 | blast007 | I copied over the handler cxx/h, and added the HAVE_ARES_LIBRARY_INIT and BUILD_ARES to config.h |
| 00:31.47 | blast007 | I'll see how it goes |
| 00:33.05 | JeffM | I don't know if build_ares does anything, that sounds like a flag for configure |
| 00:33.08 | JeffM | cus we build it every time |
| 00:33.24 | JeffM | yeah it won't build |
| 00:33.28 | JeffM | there are some changes |
| 00:34.44 | blast007 | BUILD_ARES just changes the #include for the ares.h header |
| 00:34.49 | JeffM | ahh |
| 00:34.59 | blast007 | in AresHandler.h |
| 00:35.07 | JeffM | the new areshandler dosn't take a paramater where the old one does |
| 00:35.10 | JeffM | nethandler has issues |
| 00:35.19 | blast007 | yep, noticed that too |
| 00:36.27 | JeffM | I'd ratehr not replace nethandler, pick up too much other crap |
| 00:38.21 | a_meteorite | Did anyone have any information on message protection that may be useful? Spent a few hours the other night going through commit logs searching for message, msg, protect, etc trying to find what the message protection in 2.99 *is* so I can identify the responsible commits |
| 00:38.47 | a_meteorite | Looked at lots of commits, couldn't find anything that was blaringly obvious... |
| 00:39.22 | JeffM | oh it won't hav those terms |
| 00:39.39 | JeffM | do a blame on the file that has the big message switch statement |
| 00:39.44 | JeffM | see when they got changed |
| 00:40.00 | JeffM | it wasn't one fix |
| 00:40.12 | a_meteorite | looks like bzfsClientMessages |
| 00:40.24 | JeffM | it was a lots of little fixes in responce to that guy that was spamming everyone and not joining |
| 00:40.35 | JeffM | we may have already fixed it in 2.0.x |
| 00:40.35 | a_meteorite | ah, that fix |
| 00:41.05 | a_meteorite | we don't want to move the class-handling system for messages to 2.3 do we? (a la bzfsClientMessages) |
| 00:41.11 | a_meteorite | that's pretty invasive as far as I can see |
| 00:41.11 | JeffM | oh god no |
| 00:41.16 | JeffM | that's buffering |
| 00:41.20 | a_meteorite | ah |
| 00:41.44 | JeffM | just make sure that no messages other then the required ones are accepted untill you have joined |
| 00:41.51 | a_meteorite | thought so, so basically find the sanity checks and stuff in there and make sure they're all there for 2.3 |
| 00:41.52 | JeffM | like you can't send a chat message before you login |
| 00:41.55 | a_meteorite | k |
| 00:41.56 | JeffM | yes |
| 00:43.10 | BulletCatcher | That seems to be the same thing as the "fix the protocol bug that version 2.0.16 was a band-aid for" item. |
| 00:43.17 | JeffM | could be |
| 00:56.31 | Thumper_ | BulletCatcher: the LaserMania maps on ports 5154 and 10004 are identical for 2.0 and 2.3 |
| 00:56.57 | Thumper_ | -set _wallHeight 16.56 is in the map file |
| 00:59.26 | BulletCatcher | Huh. I get "-set _wallHeight 0" when I save the map with 2.0. |
| 01:00.36 | BulletCatcher | And Louman's original R10.3 map has "-set _wallHeight 0" in it. |
| 01:04.48 | mdskpr_ | oof, did someone forget to reset the protocol version.....joining Lasermania just now really killed.... |
| 01:05.30 | BulletCatcher | That was you, mdskpr_. I was just about to discuss that with you. |
| 01:05.49 | mdskpr_ | me? |
| 01:06.02 | blast007 | mdskpr_: "really killed" isn't very descriptive |
| 01:06.34 | mdskpr_ | it made my computer 1000x slower untill i killed it |
| 01:06.58 | blast007 | so the client went into an infinite loop? |
| 01:07.09 | mdskpr_ | idk |
| 01:07.10 | BulletCatcher | You're lucky. I had to wait until mine exceeded the virtual memory limit and was killed by the system. |
| 01:07.41 | blast007 | mdskpr_: running with any modifications in the code? |
| 01:07.54 | BulletCatcher | You were running a client with your object ricochet patch, weren't you? |
| 01:07.56 | mdskpr_ | ahhh...true...forgot about that...duh |
| 01:09.07 | JeffM | blast007, I got it down to link errors in bzfs and bzflag.exe and that's it |
| 01:09.11 | BulletCatcher | You did a fine job in backporting the ricochet part of r18912. You merely overlooked the need to bump the protocol number. |
| 01:09.24 | JeffM | for some reason it can't find ares |
| 01:09.46 | mdskpr_ | yeah i wasnt quite sure how to do that. |
| 01:09.53 | JeffM | add one to the number |
| 01:09.59 | mdskpr_ | in buildDate.cxx? |
| 01:10.03 | blast007 | yep |
| 01:10.22 | BulletCatcher | A routine protocol bump during development only requires an increment of BZ_PROTO_VERSION and BZ_REV in buildDate.cxx and on the AC_INIT line in configure.ac. |
| 01:11.09 | mdskpr_ | AC_INIT only has the majore version: 2.3.1 bump the 1->2? |
| 01:11.18 | BulletCatcher | Yup. |
| 01:12.05 | BulletCatcher | Don't worry about updating the tracker patch, mdskpr_. In a moment I will give you access to commit it directly. |
| 01:13.10 | mdskpr_ | k |
| 01:13.38 | BulletCatcher | First, you need to update your Subversion config file as described at http://my.bzflag.org/w/Auto-props . |
| 01:16.50 | mdskpr_ | done |
| 01:17.44 | BulletCatcher | k |
| 01:17.48 | BulletCatcher | More generally, you should familiarize yourself with the guidelines in the DEVINFO file at the top level of the source tree. |
| 01:17.52 | BulletCatcher | The rico patch is fine in that way. |
| 01:20.21 | BulletCatcher | You now have commit access. |
| 01:20.22 | BulletCatcher | Be careful, and check with other developers prior to each of your first few commits until you feel more comfortable. |
| 01:20.34 | mdskpr_ | ok |
| 01:21.12 | BulletCatcher | For the rico update in particular, be careful not to mix in any other changes you may have. |
| 01:21.56 | mdskpr_ | i keep separate source trees for each revision |
| 01:22.32 | BulletCatcher | That's more careful than most of us are, I think. :-) |
| 01:27.21 | JeffM | blast007, I have a fix... it's not as awesome as I want but it works |
| 01:27.29 | blast007 | ok |
| 01:27.36 | JeffM | have to make ares a dll |
| 01:27.41 | CIA-116 | bzflag: 03mudskipper * r21491 10/trunk/bzflag/ (53 files in 7 dirs): Added ricochet for a per obstacle basis(for .bzw files) From trepans patch: r18912 |
| 01:27.42 | blast007 | hmm |
| 01:30.03 | BulletCatcher | mdskpr_: You missed the increment of BZ_REV in buildDate.cxx . |
| 01:30.55 | mdskpr_ | eek....thought it was only the first one |
| 01:31.14 | blast007 | first one? |
| 01:31.21 | BulletCatcher | No problem, just make another commit. |
| 01:31.22 | blast007 | 2.3.1 is major.minor.revision |
| 01:31.34 | mdskpr_ | *line |
| 01:31.41 | CIA-116 | bzflag: 03JeffM2501 * r21492 10/trunk/bzflag/ (12 files in 6 dirs): |
| 01:31.41 | CIA-116 | bzflag: only init ares once and clean it up when we go home |
| 01:31.41 | CIA-116 | bzflag: make ares a dll on windows, it just works better that way for some reason.... |
| 01:31.50 | JeffM | every time you make it incompatable bump that number ;) |
| 01:32.03 | CIA-116 | bzflag: 03mudskipper * r21493 10/trunk/bzflag/src/date/buildDate.cxx: bumped protocol version from 201->202 |
| 01:32.07 | JeffM | blast007, I was able to connect to a server with that |
| 01:32.12 | blast007 | ok |
| 01:32.25 | blast007 | I'm still gonna poke around at it some more |
| 01:32.38 | blast007 | seems like it should be able to work without a DLL, since 2.99.x can.. |
| 01:32.42 | JeffM | I think the DLL works better because it's al lextern C crap |
| 01:32.46 | JeffM | yeah that I don't know |
| 01:32.52 | JeffM | could jsut be wierd project settings |
| 01:33.04 | JeffM | cus it compiled and should have worked as a static lib, it just would not link |
| 01:33.13 | blast007 | yeah, the 2.99.x handler doesn't do extern "C" |
| 01:33.14 | JeffM | the main fix is the init stuff |
| 01:33.22 | blast007 | ok |
| 01:33.23 | BulletCatcher | ~mdskpr++ |
| 01:33.25 | JeffM | yeah i took that out too |
| 01:33.31 | JeffM | the header looks similar now |
| 01:33.49 | JeffM | maybe I've just got crufty cache |
| 01:33.58 | blast007 | sounds like I might be at a similar point then - getting the linker errors now :) |
| 01:34.05 | JeffM | heh |
| 01:34.22 | JeffM | perhaps there are other changes in net and the exes that prevent the link in 2.0.x |
| 01:34.25 | BulletCatcher | Also, mdskpr_, please add a line to the ChangeLog file noting the new ricochet feature. |
| 01:34.46 | JeffM | mdskpr_, you gooing to do the window thing too? |
| 01:35.00 | mdskpr_ | yup |
| 01:36.06 | CIA-116 | bzflag: 03mudskipper * r21494 10/trunk/bzflag/ChangeLog: updated changelog |
| 01:36.20 | JeffM | now to really break compatability ;) |
| 01:36.38 | JeffM | gotta remember how I did this |
| 01:36.45 | blast007 | ooo, the HTTP connect style? :) |
| 01:36.49 | JeffM | yeah |
| 01:37.02 | JeffM | client connects first :) |
| 01:37.25 | mdskpr_ | JeffM: would removing geometry solicit a bump in protocol? |
| 01:37.30 | BulletCatcher | In the future, mdskpr_, we like commit messages to be a bit more descriptive than just saying that a file has been updated. |
| 01:37.34 | JeffM | no |
| 01:38.04 | mdskpr_ | k |
| 01:38.18 | JeffM | the proto changes when your change makes it so other clients will ether behave differently then yours, or you change a network message |
| 01:38.27 | JeffM | cus in that case they can't parse our message |
| 01:38.47 | JeffM | the rico change may have seemed client only but it changed how the shots behaved |
| 01:38.53 | JeffM | and everyone has to have those work the same |
| 01:39.18 | JeffM | local window settings never go over the wire and everyone has different ones ayway |
| 01:45.58 | JeffM | oh someone has to do the config file upgrade too |
| 01:46.03 | JeffM | new dir etc.. |
| 01:46.22 | BulletCatcher | Client config files are already in a 2.4 subdirectory. |
| 01:46.38 | JeffM | yeah but it shold init from the 2.0 dir |
| 01:46.45 | JeffM | and copy whatever settings it can |
| 01:46.57 | JeffM | we did that from 1.10 to 2, and from 2. to 3 |
| 01:47.03 | JeffM | we are super nice that way :) |
| 01:48.58 | BulletCatcher | I'll add that task to the wiki page. |
| 01:49.07 | BulletCatcher | (But I'm not signing up for it.) |
| 01:50.03 | JeffM | thanks |
| 01:50.40 | JeffM | man I really ripped networking a new one didn't I? |
| 01:51.44 | CIA-116 | BZFlag: 03Bullet Catcher 07http://my.bzflag.org * r7652 10/w/BZFlag_2.3: /* New Changes */ import 2.0 config settings |
| 01:51.51 | BulletCatcher | You mean in 2.99? |
| 01:51.55 | JeffM | yeah |
| 01:52.03 | BulletCatcher | Yup. |
| 01:52.10 | JeffM | it's way easier to read :) |
| 01:52.25 | mdskpr_ | what's the difference between bzflag.6 and bzflag.6.in? |
| 01:53.01 | blast007 | JeffM: I might be closer on this ares issue - found a preprocessor definition that sounds useful: CARES_STATICLIB |
| 01:53.10 | blast007 | building right now and I'll see if that does it ;) |
| 01:53.18 | JeffM | that was in the old ares project |
| 01:53.25 | JeffM | I took it out to turn it into a DLL |
| 01:53.50 | BulletCatcher | mdskpr_: .6.in is the input file that is processed to produce the .6 file. |
| 01:53.56 | JeffM | ahh yeah if it's not in the people that use it they will try to load it extern |
| 01:53.57 | blast007 | yeah, but it was in net, bzfs, ares, and libcurl in 2.99.x |
| 01:54.03 | JeffM | throw it in config.h ;) |
| 01:54.15 | *** join/#bzflag jcp (alex@bzflag/contributor/javawizard2539) |
| 01:54.24 | JeffM | if BUILD_ARES set it to 1 |
| 01:54.27 | blast007 | yeah, well there's that option too :) |
| 01:54.38 | blast007 | I'll see if I get closer this way |
| 01:54.41 | JeffM | then we can just take my code and turn off the dll |
| 01:54.56 | blast007 | ok |
| 01:55.08 | JeffM | did you global init and free ares in the rest of bzflag and bzfs? |
| 01:55.24 | blast007 | I'm not sure ;) |
| 01:55.31 | JeffM | I know I did |
| 01:55.35 | blast007 | I copied some code from the 2.99.x areshandler |
| 01:55.42 | JeffM | that's not all of it |
| 01:55.52 | JeffM | there was code in other places that called that stuff |
| 01:56.01 | blast007 | I'll probably just svn up in a bit here |
| 01:56.52 | JeffM | I am trying with adding CARES_STATICLIB to config.h |
| 01:57.10 | blast007 | yay, I made bzflag crash on start ;) |
| 01:57.10 | JeffM | and turning it back into a static lib |
| 01:57.36 | JeffM | yeah you have to change the global ares to a pointer then init it |
| 01:57.42 | JeffM | or you are using bad memory |
| 01:58.22 | blast007 | bad copy/paste |
| 02:01.46 | JeffM | yeah that links, let me fix it and you can test it, I'm in the middle of chaning the connection proto so I can't join |
| 02:02.00 | blast007 | ok |
| 02:03.12 | CIA-116 | bzflag: 03JeffM2501 * r21495 10/trunk/bzflag/ (3 files in 2 dirs): link ares static again. |
| 02:03.19 | JeffM | give that a whirl |
| 02:03.33 | JeffM | while I ponder how to do http and go to the store |
| 02:07.31 | mdskpr_ | compile error>> AresHandler.cxx:90: error: invalid conversion from 'void (*)(void*, int, hostent*)' to 'void (*)(void*, int, int, hostent*)' |
| 02:07.32 | mdskpr_ | << |
| 02:08.05 | BulletCatcher | Is that on Windows or Linux? |
| 02:08.10 | mdskpr_ | Linux |
| 02:09.34 | JeffM | you probably didn't update linux to use the new ares |
| 02:09.37 | JeffM | so it's using the old one |
| 02:09.39 | JeffM | c-ares |
| 02:09.49 | BulletCatcher | Yeah, I haven't done that yet. |
| 02:10.06 | JeffM | linux work is going to be hard untill the build system is updates |
| 02:10.41 | JeffM | install windows mdskpr_ it works much nicer ;) |
| 02:10.47 | Thumper_ | :-P |
| 02:13.46 | BulletCatcher | For now, non-Windows devs may need to go no higher than r21491. |
| 02:13.47 | BulletCatcher | I'll need some time to at least find a temporary work-around. |
| 02:14.16 | Thumper_ | rewinds to r21453 |
| 02:20.57 | mdskpr_ | can we still commit changes? |
| 02:21.23 | BulletCatcher | Technically, yes, but it may be difficult to test them adequately. |
| 02:21.40 | Thumper_ | I'm going to sit on mine until I can test it on the last commit on trunk again |
| 02:22.00 | BulletCatcher | I hope to have a workaround within 1/2 hour. |
| 02:23.02 | BulletCatcher | I recommend *playing* BZFlag while you wait. :-) |
| 02:27.45 | blast007 | JeffM: woot, that worked |
| 02:33.48 | CIA-116 | bzflag: 03bullet_catcher * r21496 10/trunk/bzflag/man/Makefile.am: Remove deleted file makeHTML.bat from the distribution list. |
| 02:33.50 | *** join/#bzflag Guest632 (4aa77ed6@gateway/web/freenode/ip.74.167.126.214) |
| 02:33.57 | BulletCatcher | mdskpr_: r21496 builds successfully for me, probably because it uses the ares development package that is installed on my Linux system. |
| 02:34.16 | mdskpr_ | what version of ares is that? |
| 02:34.56 | BulletCatcher | I have c-ares-1.7.3 as shipped with Fedora 14. |
| 02:35.32 | mdskpr_ | i have 1.5.3, will upgrade |
| 02:35.55 | BulletCatcher | Yeah, that's pretty old. |
| 02:42.41 | meem_afk | r21496 Builds fine for me using Ubuntu. |
| 02:42.48 | Thumper_ | BulletCatcher: I didn't have c-ares-dev installed, after installing 1.7.3 on my system the build w/o plugins works for me too |
| 02:42.59 | Thumper_ | it fails without it installed for me |
| 02:43.30 | mdskpr_ | ok, now it works :) |
| 02:43.34 | BulletCatcher | yay |
| 02:43.40 | blast007 | yep, the non-Windows builds aren't set to use the bundled c-ares |
| 02:43.51 | BulletCatcher | configure will choose the system c-ares if the development package is available. |
| 02:49.14 | BulletCatcher | For those following along at home, the workaround is to install the c-ares development package and then run configure again. |
| 02:49.55 | meem_afk | I can tell you that I don't guarantee it, but I don't remember installing c-ares dev package but it still worked for me on Ubuntu. |
| 02:50.01 | Thumper_ | is having difficulty finding that for joevano's server |
| 02:52.34 | Thumper_ | so my test servers won't be building until that gets fixed |
| 02:53.06 | BulletCatcher | meem_afk: That was probably with version 21490 or earlier. |
| 02:54.09 | meem_afk | BulletCatcher: Nope, that was with 21496. Like I said though, I might have installed c-ares earlier for something else. |
| 02:56.23 | CIA-116 | bzflag: 03mudskipper * r21497 10/trunk/bzflag/ (7 files in 5 dirs): Removed -3Dfx, -no3Dfx. Moved -geometry <geometry-spec> to -window <geometry-spec>. |
| 03:00.34 | CIA-116 | bzflag: 03mudskipper * r21498 10/trunk/bzflag/src/platform/WinDisplay.cxx: remove 3dfx from WinDisplay.cxx also |
| 03:00.45 | *** part/#bzflag mdskpr_ (~mdskpr@108.25.108.121) |
| 03:01.04 | meem1029 | I'm going to go try to build it in my Ubuntu VM that I haven't installed c-ares on. |
| 03:40.00 | JeffM | just saved science! |
| 03:41.49 | meem1029 | Yay! |
| 03:41.55 | meem1029 | What does that mean anyway? |
| 03:43.34 | McYukon | That means he either blew something up or melted a 12" hole into a metal table |
| 03:43.39 | JeffM | that me and my partner in science scaved it |
| 03:43.52 | JeffM | huge sucsess |
| 03:47.23 | meem1029 | Hmm, in my Debian vm when I try to run r21498 I get: |
| 03:47.35 | meem1029 | ./src/bzflag/bzflag: error while loading shared libraries: libcares.so.2: cannot open shared object file: No such file or directory |
| 03:57.53 | BulletCatcher | That confirms that you don't have c-ares installed. |
| 03:58.51 | meem1029 | But I just installed it. It wouldn't even make successfully before I installed it. |
| 03:59.09 | BulletCatcher | Did you run configure again after installing it? |
| 04:02.53 | meem1029 | I think so. I'll go try it again. |
| 04:03.48 | Thumper_ | can someone try to connect to the bzflag3 test servers (any server). I've reverted the ares commits locally and want to see if this works or not |
| 04:05.45 | meem1029 | I'm connected fine. |
| 04:06.20 | meem1029 | on the BRL.ARPA.NET server |
| 04:06.21 | Thumper_ | meem1029: ok thanks. my client segfaults as soon as it tries to join and wanted to make sure it was just my client that is hosed |
| 04:06.31 | Thumper_ | um no wrong server |
| 04:06.37 | BulletCatcher | BRL is running 2.3.1, not 2.3.2. |
| 04:06.49 | Thumper_ | I think I have servers up on r21498 |
| 04:07.16 | meem1029 | Ah, maybe my client isn't the most up to date. Is bzbattleground still on 2.3.1 too? |
| 04:07.29 | meem1029 | That would explain why most the servers vanished. |
| 04:07.29 | McSpider | Thumper_: mine does that to, segfault |
| 04:07.35 | Thumper_ | :/ |
| 04:08.07 | Thumper_ | I get the same behaviour with a local server |
| 04:08.27 | meem1029 | I'm updating mine to the newest release now to try it. |
| 04:10.35 | Thumper_ | bzadmin can connect |
| 04:11.54 | Thumper_ | bzflag seems to segfault immediately in AresHandler::queryHost |
| 04:15.03 | meem1029 | Yep, I segfault now too. |
| 04:17.26 | Thumper_ | if I revert JeffM's c-ares commits I can connect again |
| 04:19.00 | meem1029 | That was r21498, right? I think I was on 21497 before when I was able to connect fine. That or 21496. |
| 04:20.13 | Thumper_ | I reverted the changes in r21495 and r21492, the servers are on r21498 |
| 04:20.40 | Thumper_ | but r21495 seems to be windows only changes so I probably didn't need that |
| 04:21.54 | Thumper_ | I'll look at this more tomorrow. gotta sleep - cya |
| 04:22.05 | Thumper_ | er today |
| 04:27.51 | Thumper_ | btw those servers report the revision number they are at on the server list |
| 04:32.31 | BulletCatcher | I get the client segfault even when connecting to localhost, using my system ares and curl libraries. |
| 04:32.51 | BulletCatcher | That is at r21498. |
| 04:33.31 | Thumper_ | ditto - I think the ares global init stuff JeffM put in isn't initializing the pointer correctly - so we're deferencing null |
| 04:33.56 | DTRemenak | Thumper_: we had the same problem in dev when those changes went in. look through the logs. |
| 04:34.05 | DTRemenak | it all got fixed up |
| 04:37.00 | Thumper_ | DTRemenak: which logs? you mean the commits on the old 2.99 trunk? |
| 04:37.05 | DTRemenak | yeah |
| 04:37.29 | DTRemenak | iirc there were changes required to bzflag.cxx/bzfs.cxx |
| 04:37.37 | Thumper_ | ok - I'll look at it tomorrow |
| 04:37.47 | DTRemenak | just look at the commits from around the same time as all the areshandler fixes, they should be pretty obvious |
| 04:38.05 | Thumper_ | ok |
| 04:38.28 | BulletCatcher | I am looking now. r21149 may be in the right timeframe. |
| 04:39.19 | BulletCatcher | Or maybe r17043. |
| 05:13.15 | BulletCatcher | Trying to solve this directly, gdb gives me: |
| 05:13.17 | BulletCatcher | AresHandler::queryHost (this=0x0, hostName=0x937c62 "localhost") at AresHandler.cxx:96 |
| 05:14.04 | BulletCatcher | I suspect that "this=0x0" means the class has not been properly constructed. |
| 05:14.17 | DTRemenak | correct, go further up the callstack |
| 05:14.46 | BulletCatcher | one up, the stack trace is in playingLoop () at playing.cxx:6180 |
| 05:15.02 | DTRemenak | what's that line? |
| 05:15.30 | BulletCatcher | ares->queryHost(startupInfo.serverName); |
| 05:15.31 | BulletCatcher | So "ares" is not initialized. |
| 05:15.37 | DTRemenak | yup |
| 05:16.03 | DTRemenak | 2.99 and grep should tell you how/when that should be done |
| 05:16.14 | BulletCatcher | Yeah. I'll get it fixed. |
| 05:17.43 | DTRemenak | probably a parallel fix needed in bzfs |
| 05:18.28 | BulletCatcher | Not obviously. |
| 05:18.38 | DTRemenak | k |
| 05:20.59 | BulletCatcher | grrr, the call to initGlobalAres() is inside #ifdef _WIN32 |
| 05:22.36 | DTRemenak | in 2.99? |
| 05:22.44 | BulletCatcher | in 2.3 |
| 05:22.50 | DTRemenak | ah |
| 05:24.00 | BulletCatcher | It is working now. |
| 05:24.01 | BulletCatcher | I just need to decide how much to backport. |
| 05:24.29 | DTRemenak | the ares changes were almost entirely done for stability reasons |
| 05:24.37 | DTRemenak | likewise with the curl changes |
| 05:24.46 | DTRemenak | nethandler not so much :) |
| 05:31.03 | CIA-116 | bzflag: 03bullet_catcher * r21499 10/trunk/bzflag/src/bzflag/bzflag.cxx: Initialize ares on all platforms. |
| 05:31.31 | BulletCatcher | It is possible, but unlikely based on 2.99, that that will cause a problem on Windows. |
| 05:33.53 | BulletCatcher | The old CIA had our project name as "BZFlag" rather than "bzflag". |
| 05:34.06 | meem1029 | Hmm, I'm getting server not found now. |
| 05:37.43 | BulletCatcher | Does it show the MOTD at startup? |
| 05:38.33 | meem1029 | ya |
| 05:39.01 | meem1029 | (This is on Ubuntu btw) |
| 05:39.10 | BulletCatcher | Then hostname lookup and connections are working. |
| 05:39.30 | BulletCatcher | Try selecting a different server from the list. |
| 05:40.28 | meem1029 | I haven't tested it with all of them, but I have with most. |
| 05:40.54 | BulletCatcher | I need a few more minutes to finish rebuilding. |
| 05:41.32 | meem1029 | And I'm pinging the server fine (they're all from bzflag3.norang.ca) |
| 05:41.39 | BulletCatcher | Single-core CPUs don't cut it anymore. |
| 05:45.19 | BulletCatcher | I connected to LaserMania. It works for me. |
| 05:46.43 | meem1029 | Hmm, it's not working for me. What platform are you on? |
| 05:46.57 | BulletCatcher | Fedora 14 Linux. |
| 05:47.12 | meem1029 | Ok. |
| 05:48.12 | meem1029 | I'll try a clean make while I go shower and get back to you in 10-15 minutes as to if it works. |
| 05:48.41 | BulletCatcher | I suggest doing a "make maintainer-clean", then starting with autogen.sh. |
| 05:49.01 | BulletCatcher | That'll clean up everything very thoroughly. |
| 05:49.06 | meem1029 | ok, sounds good |
| 05:57.26 | meem1029 | Hmm, still having the issues. I'm too tired right now to bother figuring out what the problem is though. Maybe I'll be more motivated tomorrow. |
| 05:59.25 | meem1029 | Hmm, that's interesting. I just decided to give it one last try by using the ip address as given by my ping and that worked, but by domain name still doesn't. |
| 06:02.59 | BulletCatcher | That eliminates several possibilities, but I'm not sure what remains. |
| 06:15.52 | *** join/#bzflag Marzipan- (~Marzipan@91.34.103.7) |
| 06:16.37 | *** join/#bzflag tupone (~alfredo@tupone.it) |
| 06:16.37 | *** join/#bzflag tupone (~alfredo@gentoo/developer/tupone) |
| 06:16.37 | *** mode/#bzflag [+v tupone] by ChanServ |
| 06:17.17 | *** join/#bzflag Marzipan (~Marzipan@bzflag/player/Marzipan) |
| 07:23.55 | *** join/#bzflag FastLizard4|zZzZ (fastlizard@wikipedia/pdpc.active.FastLizard4) |
| 08:01.42 | *** join/#bzflag Constitution (~const@bzflag/developer/Constitution) |
| 08:52.43 | CIA-116 | bzflag: 03cobrafast * r21500 10/trunk/bzflag/src/bzflag/HUDRenderer.cxx: Fix showCoordinates display bug. Based on patch 3302160 by mrapple. |
| 10:08.42 | *** join/#bzflag mdskpr_ (~mdskpr@108.25.108.121) |
| 10:15.44 | blast007 | BulletCatcher: I'll fix our SVN hook |
| 10:16.21 | blast007 | guess we had used the other CIA one that lets you pick a custom name |
| 11:02.34 | *** join/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 11:05.00 | *** join/#bzflag tupone (~alfredo@gentoo/developer/tupone) |
| 11:05.00 | *** mode/#bzflag [+v tupone] by ChanServ |
| 11:33.36 | *** join/#bzflag KTL2 (~KTL@213.219.156.22.adsl.dyn.edpnet.net) |
| 11:38.46 | *** join/#bzflag catay (~smertens@kaiya.catay.be) |
| 11:52.32 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 12:07.17 | joevano | Thumper_: did you find c-ares? |
| 12:31.01 | Gabe_G23 | pft http://gizmodo.com/5802027/cats-in-tanks-wreak-pure-havoc |
| 12:45.49 | KTL2 | miauw |
| 12:46.01 | *** part/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 12:47.15 | blast007 | Gabe_G23: a warning about the profanity/violence would be nice |
| 12:47.18 | blast007 | :P |
| 14:10.17 | *** join/#bzflag bier|tp (~bier@p54A5B825.dip.t-dialin.net) |
| 14:45.38 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 14:45.38 | *** mode/#bzflag [+v spldart] by ChanServ |
| 15:00.26 | *** join/#bzflag eTangenT (~lamer@c-98-245-170-238.hsd1.co.comcast.net) |
| 15:14.13 | *** join/#bzflag jh` (~jh_@212-149-234-105.bb.dnainternet.fi) |
| 15:21.37 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 15:47.00 | *** join/#bzflag Upsetter (~er@i59F6BFF4.versanet.de) |
| 15:59.35 | *** join/#bzflag WarPig (~j-mo@207.235.53.82) |
| 16:14.19 | *** part/#bzflag ts (~ts@bzflag/developer/ts) |
| 16:14.29 | *** join/#bzflag ts (~ts@bzflag/developer/ts) |
| 16:14.39 | ts | arrgs |
| 16:15.01 | ts | that's what I get for using arrow keys |
| 16:16.21 | ts | wanted to say someone broke the build process..AresHandler.h not finding ares.h |
| 16:25.46 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 16:25.46 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 16:26.04 | blast007 | ts: do you have a system installed copy of the ares development files? |
| 16:26.22 | blast007 | currently the build system on *nix doesn't properly utilize the ares in src/other/ |
| 16:27.09 | ts | blast007: no, I have no ares lib installed in system AFAIK |
| 16:38.21 | *** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501) |
| 16:38.21 | *** mode/#bzflag [+v JeffM] by ChanServ |
| 16:39.54 | *** join/#bzflag jh`` (~jh_@212-149-234-105.bb.dnainternet.fi) |
| 18:16.02 | blast007 | ts: did it work after you installed the ares dev package? |
| 18:16.04 | *** join/#bzflag Cobra_Fast2 (~coprah@koln-4d0b1b97.pool.mediaWays.net) |
| 18:17.30 | *** join/#bzflag Erroneous (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 18:17.30 | *** mode/#bzflag [+v Erroneous] by ChanServ |
| 19:36.13 | CIA-116 | BZFlag: 03JeffM2501 * r21501 10/trunk/bzflag/MSVC/build/bzfs.sln: server only project to help with the debugin |
| 19:48.54 | *** join/#bzflag kierra (~jolie@unaffiliated/kierra) |
| 20:17.05 | Gabe_G23 | blast007: oh yeah, forgot about that sorry :P |
| 20:34.02 | *** join/#bzflag a_meteorite (~bzmeteori@unaffiliated/ameteorite/x-000000001) |
| 20:37.09 | *** join/#bzflag CIA-61 (cia@cia.atheme.org) |
| 20:46.57 | *** join/#bzflag CIA-114 (cia@cia.atheme.org) |
| 20:46.59 | *** join/#bzflag KTL (~KTL@213.219.156.22.adsl.dyn.edpnet.net) |
| 21:14.23 | *** join/#bzflag sirquine (~quine@c-67-186-15-145.hsd1.pa.comcast.net) |
| 21:15.41 | JeffM | the AC_INIT is where you have to bump the rev in configure.ac right? |
| 21:19.38 | CIA-114 | BZFlag: 03JeffM2501 * r21502 10/trunk/bzflag/ (13 files in 7 dirs): |
| 21:19.38 | CIA-114 | BZFlag: Have client send connection header to server to be more like HTTP and other interweb protocols. Use connection header of "BZFLAG\r\n\r\n" |
| 21:19.38 | CIA-114 | BZFlag: make the networking use tubes instead of trucks. |
| 21:19.38 | CIA-114 | BZFlag: bump rev and proto. |
| 21:19.38 | CIA-114 | BZFlag: note. tools that connect to servers to check stats need to be changed to use the "v3" method. |
| 21:19.58 | JeffM | compatability --------------> window --------> |
| 21:25.59 | meem1029 | Hmm, I'm still getting problems connecting to the server in r21502. |
| 21:26.16 | JeffM | that's probably ares |
| 21:26.26 | JeffM | I did nothing to fix the linux build issues |
| 21:26.36 | meem1029 | Oh, ok. |
| 21:27.04 | JeffM | I have no desire or method to work on the linux build system |
| 21:27.19 | Thumper_ | JeffM: :) the init of ares was inside a #ifdef windows |
| 21:27.28 | JeffM | yeah I saw, that was my bad |
| 21:27.35 | JeffM | good catch |
| 21:28.48 | JeffM | I don't expect linux to work very well untill BulletCatcher finishes moving over the build system from 2.99 |
| 21:28.54 | JeffM | and it's a large task |
| 21:29.08 | JeffM | harder then doing the windows one |
| 21:29.11 | Thumper_ | we'll limp along in the meantime :) |
| 21:29.17 | JeffM | since all my files are in one spot |
| 21:31.39 | CIA-114 | BZFlag: 03JeffM2501 07http://my.bzflag.org * r7653 10/w/BZFlag_2.3: /* Backports */ add push stats |
| 21:40.22 | McSpider | Mac OSX, bzflag r21502 make error: http://pastebin.com/sW5Gzr5h |
| 21:43.13 | CIA-114 | BZFlag: 03JeffM2501 * r21503 10/trunk/bzflag/src/bzflag/ServerLink.cxx: silly #defines for people who don't have winsoc |
| 21:43.18 | JeffM | give that a shot |
| 21:44.15 | JeffM | I have no way to compile or test the code in that #def |
| 21:46.01 | McSpider | http://pastebin.com/WgvvUmQt |
| 21:46.27 | JeffM | you'll have to fix that part |
| 21:47.12 | McSpider | humm, ok I'll try |
| 21:56.18 | *** join/#bzflag Anxuiz (~Anxuiz@unaffiliated/anxuiz) |
| 21:59.45 | McSpider | I added two brackets, now it compiles. should I post a diff to the tracker? |
| 22:00.53 | JeffM | pastebin the changes reall fast |
| 22:01.38 | McSpider | http://pastebin.com/QTGeLU3r |
| 22:02.28 | JeffM | odd |
| 22:03.33 | McSpider | what? |
| 22:04.43 | JeffM | I don't undertand why putting brackets there fixes it |
| 22:05.11 | McSpider | I found this when I googled the error: http://www.karakas-online.de/forum/viewtopic.php?t=4194 |
| 22:05.19 | McSpider | It explains why |
| 22:05.34 | JeffM | oh cus of the goto.. |
| 22:05.35 | JeffM | hmmm |
| 22:05.40 | JeffM | I removed the goto on windows |
| 22:05.45 | JeffM | probably don't need it on linux ether |
| 22:09.47 | CIA-114 | BZFlag: 03JeffM2501 * r21504 10/trunk/bzflag/src/bzflag/ServerLink.cxx: remove the goto calls.. they are evil. |
| 22:09.48 | JeffM | give that a whirl |
| 22:10.28 | JeffM | the goto was saving one line per call, not a big thing |
| 22:10.41 | JeffM | and oen was simply replaced by inverting an if :) |
| 22:11.16 | McSpider | works fine, and I agree goto's are evil |
| 22:31.22 | CIA-114 | BZFlag: 03JeffM2501 * r21505 10/trunk/bzflag/ (7 files in 2 dirs): add API hooks for non player connections (HTTP). |
| 22:31.51 | JeffM | didn't bother with IRC connections this time |
| 22:32.18 | JeffM | and only non player connections go thru the new code after the header is sent, so it should be a lot less risky then what was in 2.99 |
| 22:34.10 | blast007 | cool |
| 22:34.55 | JeffM | haven't done the rest of the API yet but we should be able to get the HTTP plugins in when I do |
| 22:57.59 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 23:06.00 | *** join/#bzflag catay (~smertens@kaiya.catay.be) |
| 23:06.06 | CIA-114 | BZFlag: 03Mrapple 07http://my.bzflag.org * r7654 10/w/Web_Services: Add link for home page |
| 23:14.59 | *** join/#bzflag meem1029 (~meem1029@2001:468:1930:24:7add:8ff:fee3:6a82) |
| 23:37.38 | *** join/#bzflag Guest92489 (~trepan@204.237.8.232) |
| 23:49.32 | mdskpr_ | compilation error on linux: http://pastebin.com/NWwWgDpU |