Monday, 2007-02-26

*** tpb has joined #tp00:03
*** ChanServ sets mode: +o tpb00:03
tpbtpb has joined on worldforge00:03
*** mithro has joined #tp00:08
nashheyo mithro00:12
mithrohowdy nash00:13
nashhow is life?00:13
mithrookay00:21
mithrois this where you tell me something is horrible borked?00:21
nashWell I think so... but I can't prove it yet.00:22
nashI will be able to be the end of the day most likely00:22
nashI'll be putting in the tests during the day when I have downtime ;-)00:23
nashAs I said the other day - It seems that my orders are getting screwed up sometimes.00:25
nashSo I need to prove it...00:25
mithronash: are you still sending probe orders?00:29
nashYep00:29
nashDo you want the problem description?00:29
mithroi'm actually guessing thats likely to be the cause :)00:29
nashProbably not.00:30
nashThe problem is:00:30
nasha) AI sets move order to planet00:30
nashb) Ai sets col order on planet00:30
nashResult is00:30
nasha) Turn 1-N.. ship moves to planet00:30
nashb) Get 'Moved to X' message00:30
nashc) Ship then begins to move with a {dx,dy} of either ({1,0,-1},{1,0,-1}) where dx != 0 && dy != 000:32
nashSorry where dx != 0 || dy != 0 (so one of the 8 cardinal directions)00:32
mithrois it heading for the center of the universe?00:34
nashNope - One of the distant corners00:34
mithrookay not what I was thinking then00:34
nashSo... you know this bug?00:34
nashOr a similar one?00:35
nashSo anyway what I need to do is make sure I download the full orders for objects, so I can dump them out and see WTF is going on00:36
mithronash: i was thinking that maybe the server wasn't correctly cleaning up the probe orders00:36
nashright00:36
* nash notes it's a good thing you didn't offer money for bugs... you wouldn't be able to afford it ;-)00:44
mithronash: :)00:47
mithroI guess I should start organising prizes00:53
nashYou have a month ;-)00:54
nashDo you know how many entrants there are?00:54
* nash is aware of just one more... the mac guy ;-)00:54
mithrocurrently i'm aware of ~3 including you00:57
nashHow many did you expect?01:02
JLPmithro: would this RSS feeds plugin for phpBB be ok - http://www.phpbbhacks.com/download/621201:03
tpbTitle: phpBBHacks.com - Smartfeed (at www.phpbbhacks.com)01:03
mithroi was hoping for 10, realisiticly i though about 5-601:03
nashSo short then :-(01:04
nashI thought your LCA stuff may have got a few more out of the woodwork01:04
mithroJLP: looks good then!01:04
mithrowell atleast this channel has a few more people in it01:05
nashHow many before?01:06
mithroabout 801:07
mithrobut most of these people wern't "recurited" via LCA01:07
nashSo an extra 5 or so01:09
mithrobblr01:11
nashokay01:11
mithroactually have to do some work today :)01:20
nashLife can be tough sometimes01:24
mithrowe finally have more people then #freeorion :)01:31
nashheh01:34
nashWhat is there status?01:58
nashWow - they have a nice UI02:02
jothammithro?02:05
jothamping02:05
jothamthe summary screen is gonna be a bit tricky if you can have infinite players in a game02:05
jothami could do a table i guess02:05
jothamwith a scroll bar02:05
mithrojotham: hey02:05
mithrojotham: i think work for the case of a maximum of 3-402:06
jothamaight02:06
mithrooff to lunch02:17
mithrobblr02:17
nashSounds like a plan02:17
*** tpb has joined #tp06:00
*** ChanServ sets mode: +o tpb06:00
*** llnz has joined #tp06:03
mithrohowdy llnz06:08
mithroyour in a bit early today?06:09
llnzhi mithro06:09
llnzhehe, adsl online :-)06:09
mithroyay!06:09
mithrollnz: do you think you could add the intercept order and then do a release? (not half demanding am I :)06:09
llnzhumm... i'll think about it06:10
llnzthinking of your libltdl patch too06:10
mithrollnz: i've been running it here for days and it seems to work06:10
mithrollnz: any big problems about it?06:11
llnznot really06:11
mithrollnz: btw do we do darcs backups anywhere?06:12
nashMay I request a delay of ~24 houts until I see if there is bug?06:12
llnzmithro: not that i know of06:12
llnznash: why? what bug?06:12
mithrollnz: I guess I should set that up06:12
nashThe order corruption bug06:12
nashWhich I'm trying to isolate at this moment06:13
*** llnz changes topic to "AI Programming Competition in progress - http://www.thousandparsec.net/tp/comp.php | metaserver - metaserver.thousandparsec.net | 2007-02-25, TP ranked 128th on SF"06:13
llnznash: ok06:13
nashI think I'm getting closer...06:14
nashHmmm06:14
*** JLP has quit IRC06:14
mithrollnz: i think that was because of my tailor testing :)06:15
mithrollnz: it's also going to turn out if I push to sf, all changes are going to get logged under my name06:15
mithrowell I'm heading home now06:17
mithrosee ya!06:17
llnzcya06:20
nashWell there is definitly a bug06:20
nashA fleet with a "colonise" order is moving through space06:20
llnzok06:21
nashHaven't printed the full details but the blurb looks like this:06:21
nash        OrderSlot 0 Type[ 3]: Colonise06:21
nash        Turns: 10  Resources: 006:21
nashHere is teh full details of one of htem06:29
nashObject ID 294 Type 4 Name: 'A Fleet'06:29
nash        Size 206:30
nash        { x = -1271960000, y = -1732067704, z = 0 }06:30
nash        { dx = 0, dy = -200000000, dz = 0 }06:30
nash        0 children:06:30
nash        5 orders: 3 1 0 4 506:30
nash        1 current orders06:30
nash        OrderSlot 0 Type[ 3]: Colonise06:30
nash        Turns: 10  Resources: 006:30
nashllnz: Any ideas?06:30
nashA fleet with a colonise order should not move right?06:30
llnzin tpserver-cpp, the colonise order will move the fleet to the planet first, if needed06:31
nashRight...06:31
* nash notes this should be documented... 06:31
nashHowever, AFAIK it still isn't moving to the right spot06:32
nashLook at the direction it is travelling...06:34
llnzwhere is the target system?06:37
*** mithro has quit IRC06:40
nashWhere the ship started...06:48
nashAnyway - I need to head home06:49
nashTalk to you later06:49
*** nash has quit IRC06:49
*** mithro has joined #tp07:12
*** bokey has quit IRC07:12
mithrohowdy peoples07:12
llnzmithro: do you build without mysql?07:13
mithrollnz: dunno07:13
mithroI probably build with mysql07:13
llnzbecause I've just had to add linker things to make the minisec module load07:14
mithrowith ltdl?07:14
llnzwas getting undefined tpmysql symbols07:14
llnzyes with ltdl07:14
mithrohrm...07:14
mithroI guess i'm building without mysql then :)07:14
mithrodefinately building without mysql under windows07:15
llnzi have it working now, but it spits out a great big warning message07:15
llnz*** Warning: Linking the shared library libminisec.la against the loadable module07:15
llnz*** libtpmysql.so is not portable!07:15
mithrohrm...07:16
mithrocan you commit any changes and I'll have a look?07:16
mithroout of intrest why does a ruleset module need to know about  a persistance module :)07:17
llnzbecause of object type persistence07:17
* llnz is waiting for object parameterisation07:17
mithrollnz: my guess is that it's pretty easily solved07:19
llnzhold on a moment while i recompile07:19
mithrolibminisec.so uses things from tpserver-cpp core and it doesn't complain07:19
llnzit's odd07:19
mithrohrm, this might take a while, it needs to get 19mb and i'm rate shaped07:20
llnzRTLD_GLOBAL was the flag in the old dlopen code to make the symbols available from each library07:22
mithrollnz: they should be available07:23
llnz2007-02-26 20:09:13 < Error > Failed to load plugin "../modules/games/minisec/.libs/libminisec": "../modules/games/minisec/.libs/libminisec.so.0: undefined symbol: _ZTI15MysqlObjectType" (nil)07:26
llnz2007-02-26 20:09:13 <Warning> Did not load ruleset "../modules/games/minisec/.libs/libminisec"07:26
mithroThere is no need to pass `RTLD_GLOBAL' (or equivalent) to lt_dlopen as might be necessary with the native module loading API.07:26
* llnz trys again07:27
mithroyou loaded the module libtpmysql.so?07:28
llnzyes07:28
mithroit loaded okay?07:28
llnzyes07:29
llnz2007-02-26 20:09:13 < Info  > Loading persistence method ../modules/persistence/mysql/.libs/libtpmysql07:29
llnz2007-02-26 20:09:13 < Debug > Loaded plugin "../modules/persistence/mysql/.libs/libtpmysql" sucessfully07:29
llnz2007-02-26 20:09:13 < Debug > Initialisation function for plugin "../modules/persistence/mysql/.libs/libtpmysql" found07:29
llnz2007-02-26 20:09:13 < Info  > Loaded plugin "../modules/persistence/mysql/.libs/libtpmysql" sucessfully07:29
llnz2007-02-26 20:09:13 < Info  > Loaded persistence method ../modules/persistence/mysql/.libs/libtpmysql07:29
llnz2007-02-26 20:09:13 < Info  > Loading ruleset ../modules/games/minisec/.libs/libminisec07:29
llnz2007-02-26 20:09:13 < Error > Failed to load plugin "../modules/games/minisec/.libs/libminisec": "../modules/games/minisec/.libs/libminisec.so.0: undefined symbol: _ZTI15MysqlObjectType" (nil)07:29
llnz2007-02-26 20:09:13 <Warning> Did not load ruleset "../modules/games/minisec/.libs/libminisec"07:29
mithroohh, did you add a "-export-dynamic" to libtpmysql?07:30
llnznope07:30
llnzgrrr. now its working....07:30
mithro?07:30
llnzadded "libminisec_la_LIBADD = @MYSQL_LIBS@" to minisec/Makefile.am and now it works07:31
* llnz does a "make clean all" just to check07:32
mithrollnz: well that makes a little sense07:32
llnz@MYSQL_LIBS@ are the real mysql libraries, maybe it was a chained dependancy problem07:32
mithrollnz: that is what I was thinking07:34
llnzonce object parameterisation is done, then i won't even need that07:34
llnzhumm... now it doesn't go again07:35
mithrohrm....07:35
* llnz adds --export-dynamic to tpmysql to see if it makes a difference07:40
llnznope07:41
mithrollnz: testing here now07:41
mithrollnz: it's only one -07:42
llnzoh07:42
llnzyeah, i did actually put only one in the makefile.am07:42
mithroarg, I'm getting a segfault now07:48
mithro2007-02-26 18:21:07 < Info  > This is GPL software, please see the COPYING file07:49
mithro*** glibc detected *** ./tpserver-cpp: malloc(): memory corruption: 0x081084d8 ***07:49
mithro======= Backtrace: =========07:49
mithro/lib/libc.so.6[0xb7bf5082]07:49
llnzodd07:52
mithroweird, just recompiled and not it's working07:53
mithrogetting similar error to you now07:53
mithrotim@ultraslim:/opt/tpserver-cpp/bin$ c++filt _ZTI15MysqlObjectType07:58
mithrotypeinfo for MysqlObjectType07:58
mithrohrm...07:58
mithrotpmysql doesn't seem to export MysqlObjectType08:04
mithrollnz: any luck?08:12
llnznot yet, google isn't helping much either (other than a discussion about whether libtool should do RTLD_GLOBAL or not08:13
llnz)08:14
mithrollnz: my inspection of the files using objdump appears that the .so isn't ending up with that symbol08:16
llnzhumm... ok08:16
mithrothis is very weird08:34
mithrollnz: reading the documentation - you need definatly need -export-dynamic08:40
llnzit's there, after -module08:45
* llnz trys before08:47
llnznpe08:49
llnznope08:49
*** nash has joined #tp08:52
* nash waves08:52
nashmithro: llnz: Either of you around?08:53
mithroboth08:53
nashCool08:54
nashSo I've definitly got a bug in tpserver-cpp08:54
nashI'm going to paste a dump08:54
nashObject ID 325 Type 4 Name: 'A Fleet'08:55
nash        Size 208:55
nash        { x = -2054960000, y = -1676002296, z = 0 }08:55
nash        { dx = 0, dy = 200000000, dz = 0 }08:55
nash        0 children:08:55
nash        5 orders: 3 1 0 4 508:55
nash        1 current orders08:55
nash        OrderSlot 0 Type[ 3]: Colonise08:55
nash        Turns: 21  Resources: 008:55
nashColonising object 191:08:55
nashObject ID 191 Type 3 Name: 'Illuntara 1'08:55
nash        Size 208:55
nash        { x = -2054960000, y = -1876002296, z = 0 }08:55
nash        { dx = 0, dy = 0, dz = 0 }08:55
nash        0 children:08:55
nash        0 orders:08:55
nash        0 current orders08:55
nashSo as you can see, the ship is heading away from the planet it should be colonising...08:56
llnzodd08:56
nashllnz: So - it definitly looks wrong...08:58
llnzyes08:58
nashGood... It's been driving me crazy since saturday as I tried to work out WTF was going on08:59
nashTo get a nice small dump with the problem is good08:59
mithronash: so how does this occur?08:59
nashI just send a frigate to colonise a planet, sometimes it happens09:00
nashIt seems to happen to all ships sent to the same system, or it seems to work fine.09:01
llnzhummm... i think i know what it is, maybe09:02
nashShall I report a bug for the extra stats and activity?09:04
mithrollnz: ah ha!09:08
llnzsure, it wasn't what i think it was09:08
llnzmithro: what?09:08
tpbmithro: what is the definision of an empty list - just a '0' (length field), llnz09:08
llnzhehehehe09:08
mithrollnz: Additionally, the installed version with Ubuntu 6.06 does not export symbols globally.09:08
llnzoh, so now what....09:09
llnz?09:09
mithrollnz: libltdl is designed to be shipped with your program09:09
llnzchange from installable to convience lib?09:11
mithrollnz: i'll figure out the problem and submit you a patch09:14
mithrogo figure out nash's bug :)09:14
nashSo glad delayed the release09:14
nashllnz: Any clues where to start looking for the bug?09:22
* nash can llok as well09:22
mithrocolonise.cpp?09:22
*** llnz2 has joined #tp09:29
mithro[19:46] <mithro> llnz: i'll figure out the problem and submit you a patch09:29
mithro[19:46] <mithro> go figure out nash's bug :)09:29
mithro[19:46] <nash> So glad delayed the release09:29
mithro[19:54] <nash> llnz: Any clues where to start looking for the bug?09:29
mithro[19:54] * nash can llok as well09:29
mithro[19:54] <mithro> colonise.cpp?09:29
nashWell I've founf one bug... in +66 move.cpp09:31
*** llnz has quit IRC09:32
nashreturn (int)(distance / max_speed) + 1;09:32
llnz2that's not a bug09:35
llnz2i think09:35
*** llnz2 is now known as llnz09:35
nashYes it is09:35
nashif distance == 0... time = 109:35
nashif distance == max_speed -> time = 209:36
nashOtherwise it is correct09:36
* nash just put in a bug for you.09:37
* nash may as well boost sourceforget ratings09:37
mithronash: thanks09:45
nashmithro: Also if you like tell me some info, and I can update the wiki entry09:46
nashs/wiki/wikipedia/09:47
mithronash: what info do you want?09:47
nashhave you read it?09:47
nashWhat was the project originally called?09:48
mithroThousand Parsec09:48
mithroit's always been called Thousand Parsec, before that it didn't have a name09:48
nashAccording to you email it had a different name before...09:48
nashTHe one linked from the wiki entry09:49
llnzmithro and i were helping out in the WorldForge project09:49
llnzbut it's a very different project09:50
mithronash: hrm, I can't remeber the original name - I don't think it had a proper one09:52
mithromy original announcement of Thousand Parsec is in the wf archives somewhere09:53
mithrohttp://mail.worldforge.org/archive/general/2002-January/003605.html09:55
tpb<http://ln-s.net/JKZ> (at mail.worldforge.org)09:55
mithroi think that is the earliest09:55
nashhmm09:56
nashMay wait for the AI compe09:56
nashto finish09:56
*** JLP has joined #tp10:00
llnznash: does the move order suffer the same problem as colonise?10:00
mithrohey JLP10:02
llnzhi JLP10:02
nashllnz: It could - but I haven't seen it10:02
*** _JLP_ has joined #tp10:03
mithrohello again _JLP_  :)10:04
*** JRepin has joined #tp10:06
mithroJRepin: DSL not going very well?10:10
mithronash: when Lee joined10:12
mithrohttp://mail.worldforge.org/archive/general/2002-January/003625.html10:12
tpb<http://ln-s.net/JK_> (at mail.worldforge.org)10:12
nashWell you have been quoted badly10:14
nashllnz: Is it because the moveOrder tries to do something?10:16
*** JLP has quit IRC10:17
*** _JLP_ has quit IRC10:19
*** JRepin has quit IRC10:23
*** JRepin has joined #tp10:24
mithroJRepin: you alive this time? :)10:24
*** _JLP_ has joined #tp10:28
mithroguess not10:30
*** rfhg has joined #tp10:32
mithrohe he10:33
mithrotim@ultraslim:~/oss/tp$ file ~/libltdl.patch10:33
mithro/home/tim/libltdl.patch: ASCII English text, with very long lines10:33
llnzhehe10:33
llnz> autoconf...10:35
llnz> Doing the same for libltdl10:35
llnz> cp: cannot create regular file `[.]/install-sh': No such file or directory10:35
llnz> automake: configure.ac: installing `[.]/install-sh'10:35
llnz> error while copying10:35
mithrohrm....10:35
llnzsame of mkinstalldirs missing ltmain.sh config.guess config.sub10:36
*** JLP has joined #tp10:37
mithroconfig.guess and config.sub are generated?10:37
llnznormally10:37
mithrohrm, ltmain.sh should be in the patches10:38
mithro[email protected]**20070226101038] {10:38
mithroaddfile ./libltdl/ltmain.sh10:38
mithrohunk ./libltdl/ltmain.sh 110:38
llnzi see it there10:38
mithrooh, did you rerun autogen.sh?10:39
llnzyes, no change10:39
mithrocmdline I used "sh autogen.sh && PKG_CONFIG_PATH=/opt/tpserver-cpp/lib/pkgconfig ./configure --prefix=/opt/tpserver-cpp && make clean && make && sudo make install"10:39
mithroDoing the same for libltdl10:41
mithro/usr/share/aclocal/libmcrypt.m4:17: warning: underquoted definition of AM_PATH_LIBMCRYPT10:41
mithro  run info '(automake)Extending aclocal'10:41
mithro  or see http://sources.redhat.com/automake/automake.html#Extending-aclocal10:41
mithro/usr/share/aclocal/glib.m4:8: warning: underquoted definition of AM_PATH_GLIB10:41
mithroconfigure.ac: installing `./install-sh'10:41
mithroconfigure.ac: installing `./missing'10:41
tpb<http://ln-s.net/JKd> (at sources.redhat.com)10:41
mithroconfigure.ac:69: installing `./config.guess'10:41
mithroconfigure.ac:69: installing `./config.sub'10:41
mithroremoving config.cache...10:41
*** _JLP_ has quit IRC10:46
llnzjust doesn't work for me10:47
mithroI'll submit you a new set10:47
*** rfhg has quit IRC10:48
*** JRepin has quit IRC10:50
*** JLP has quit IRC10:53
*** llnz has quit IRC10:56
mithronash: you got time to test something?10:59
nashpossibly11:02
nashwhat?11:02
*** llnz has joined #tp11:02
nashmithro: ?11:02
tpbRumor has it mithro: is this ok, nash11:02
nashmithro: what do you want me to test?11:12
mithronash: don't worry for the momen11:12
mithrot11:12
nashbtw; sf stats - now 12611:14
llnzday not finished yet, midnight UTC11:15
nashbetter report more bugs then11:17
mithrohe he11:18
nashor someone slose some11:19
nashmithro: one I reported you can close with documentation...11:19
mithroarg, it was working and now it's not :/11:20
nashwhat?11:25
mithroyay, figured it out!11:27
nashit would be what?11:28
mithronot including everyting in the pach11:29
mithrobah11:29
nashAnyway - I'm off to bed.11:31
nashLet me know if there is any news on that bug.  It's a bit of a show stopper for me.11:31
mithronash: okay11:32
llnzi'll have a look11:32
nashGood night11:32
nashllnz: thanks ;-)11:32
nashI appreciate it11:32
*** nash has quit IRC11:32
*** llnz has quit IRC12:02
*** adhoc has quit IRC12:02
*** mithro has quit IRC12:03
*** tuna-fish has quit IRC12:03
*** zipola has quit IRC12:03
*** bobbens has quit IRC12:03
*** llnz has joined #tp12:04
*** adhoc has joined #tp12:04
*** mithro has joined #tp12:04
*** tuna-fish has joined #tp12:04
*** bobbens has joined #tp12:04
*** zipola has joined #tp12:04
*** jotham has quit IRC12:05
*** jotham has joined #tp12:05
CIA-20[email protected] * tpserver-cpp/ (378 files in 11 dirs):12:13
CIA-20Make libltdl work properly on borked Debian/Ubuntu.:12:13
CIA-20 Debian and Ubuntu ship a horrible broken version of libltdl. Unlike all12:13
CIA-20 documents mention, the Debian and Ubuntu version does not export all symbols (IE12:13
CIA-20 RTLD_GLOBAL).12:13
CIA-20[email protected] * tpserver-cpp/ (24 files in 7 dirs): Working libltdl support.12:13
* llnz ponders12:13
mithrollnz: what are you pondering?12:14
mithro378 files in 11 dirs?12:14
llnzif i should try to fix the two latest bugs, or sleep12:14
llnzdarcs-cia can't count properly12:14
mithrollnz: yeah, 1 file == 1 hunk12:15
llnzyeah12:15
llnzalso, adding a file == 1 hunk == 1 file12:15
mithrollnz: how do you think we should document orders?12:17
llnzorderdesc are the documentation, but for things like that, i don't know12:17
llnzit a per-ruleset implementation thing12:18
mithroyes12:18
mithroit should probably be part of the OrderDesc12:18
mithroi think we probably need to relook into orders12:19
llnzthere is a description field in OrderDesc12:19
llnzwhat for?12:19
llnzhehe, tpsrever-cpp just go really big12:19
mithros/go/got12:19
llnz17k c++, 8k sh, 3.5k c12:20
llnzyeah12:20
llnz29048 lines all up12:20
llnzwow, over 78k lines for the project12:20
mithrohow big is libltdl?12:20
llnz8k sh, 3.5k c12:21
mithroit should probably be in an exclusion12:21
llnzyeah12:21
* llnz wanders off12:22
llnzlater all12:22
*** llnz has quit IRC12:23
*** adhoc has quit IRC13:48
*** jotham has quit IRC13:48
*** TBBle has quit IRC13:48
*** czr has quit IRC13:48
*** jasmine has quit IRC13:48
*** tuna-fish has quit IRC13:48
*** zipola has quit IRC13:48
*** mithro has quit IRC13:48
*** bobbens has quit IRC13:48
*** Demitar_ has quit IRC13:48
*** CIA-20 has quit IRC13:48
*** adhoc has joined #tp13:49
*** jotham has joined #tp13:49
*** TBBle has joined #tp13:49
*** czr has joined #tp13:49
*** jasmine has joined #tp13:49
*** CIA-20 has joined #tp13:49
*** Demitar_ has joined #tp13:49
*** mithro has joined #tp13:49
*** tuna-fish has joined #tp13:49
*** bobbens has joined #tp13:49
*** zipola has joined #tp13:49
*** adhoc has quit IRC13:49
*** adhoc has joined #tp13:49
*** JLP has joined #tp15:52
*** Demitar_ has quit IRC16:29
*** Demitar has joined #tp16:35
*** jotham has quit IRC20:30
*** tuna-fish has quit IRC20:35
*** tuna-fish has joined #tp20:39
*** nash has joined #tp21:56

Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!