Saturday, 2007-03-10

*** tpb has joined #tp00:00
*** ChanServ sets mode: +o tpb00:00
* nash waves00:01
nashmithro: heyo?00:01
*** tuna-fish has quit IRC00:58
mithronash: hey01:59
mithroyou get up a bit early01:59
nashheyo mithro02:02
nashEarly - first message was 9:45...02:02
nashAnyway did you get the AI to build before happily?02:02
mithroi only just got up then :)02:02
mithroyes, it was quite easy once I had the evas stuff02:03
nashcool02:04
nashWell if you want, the AI is playable now... Not agreesive yet, and a few stupid things02:04
nashSo if you want to see how ahrd it is to beat.. now wold be a good time02:05
nashIf you have hte time that is ;-)02:05
nashI don't know when I'm going to have time.02:05
nashI'm currently packing books into boxes for moving02:05
nashIf you still have the checkout, cg-update should do the job to get the code02:08
mithronope I removed the checkout02:09
mithroso I wouldn't be tempted to hack on it instead of my own projects02:10
nashheh02:11
nashWant the URL again again?02:11
nashcg-clone http://tranquillity.nash.id.au/tp/eclient.git02:12
mithrothanks02:13
nashJust give it a short run and see how it goes...02:14
nashAlso there is a new screenshot up for you02:14
mithromaybe I should write my own AI to give your a run for it's money :)02:14
nashYou can if you like - personally I'd prefer to see more work towards getting a MTSec release ready... so then we can both do AIs for that ;-)02:14
mithronash: yes, that was what I was thinking02:15
nashBTW: Can ships load missiles to other ships in MTSec?02:16
mithronash: probably02:16
nashif not... they should be able to so you can actually hapve supply lines as a game element.02:16
nashWhich would be cool02:16
mithroyeah, I can't see why not - maybe we should only restrict it to ships which have a "air lock" part?02:17
nashNah.02:17
nashOr maybe02:18
nashHmm...02:18
nashI could argue either wya02:18
mithroI guess maybe go the KISS route02:18
nashMaybe not "air lock" as a name.. "Space Transfer Dock/Bay"02:18
nashThat would be my main argument against02:18
nashYou can do it - standard ship fitting.02:18
nashUnless you need it to reload them at all (anywhere) so you can have 1 shot ships...02:19
mithroyes, that would be a better name02:19
mithrothe whole loadable weapons gives the defender quite a bit advantage02:20
nashAlso need to check how the transfer orders work so it doesn't take a silly amount of time to do the loading (ie One freighter shoudl be able to reload a whole fleet in a turn (since a turn is something like 1 year long))02:20
nashAhh... but an attacker can still lay seige to take out incoming supply ships.  A single planet won't be able to hold out forever02:21
mithroyes that is also true02:21
mithroscouts are most probably fast enough to "punch through" most blockades02:22
nashAlso true... However - intercept orders will help there...02:25
nashthat compile okay for you?02:31
mithronash: seem to compile okay02:39
nashcool02:39
mithronash: you need some better error handling02:40
nashMay want to tweak tpe_gui (look for connect)02:40
nashmithro: Yeah02:40
nashAnything in particular?02:40
mithrowhen you don't correct properly02:40
nashIndeed02:41
nashThat is to be fixed when I fix up the start up to not have to hardcode URLS and the like02:41
nashAlso need to add auto connect ;-)02:42
mithrohrm, it didn't like downloading the universe for some reason02:42
nashFrom where?02:43
mithroahh it's trying to connect to 10.0.0.102:43
nashAs I mentioned... it's a bit broken for connecting in general02:44
mithroyes02:44
nashThe AI itself is fine.. everything else needs work ;-)02:44
mithroyour message handler needs to handle new lines :P02:45
mithroSending Seq 23 Type 14 Len: 0 [(nil)]02:45
mithroHandling a MsgTimeRemaining (15) (response: 23)02:45
mithroSegmentation fault02:45
mithroopps02:45
nashWeird02:45
nashCan you give me a backtrace on that?02:46
mithrosure,02:47
mithroit's even reproduceable02:48
mithrojust click on the message icon then click it again02:48
mithroand why is there are 41x(null) in your home system?02:48
nashAhh...02:49
nashWell messages aren't complete yet ;-)02:49
nashAs for 41 Null - ???02:49
mithroit says "<number> - (null)" in the homesystem pop-up02:52
mithrogetting you a screenshot02:53
mithroTBBle: ping?02:57
nashFailed to transfer02:57
mithronash: email?02:57
nashwait02:58
nashworked now02:58
mithroworked?02:58
nashCan you try it again02:58
nash* DCC RECV connect attempt to mithro failed (err=Connection refused).02:59
mithro:/02:59
nashemail, or tell me where I can scp it from02:59
mithrojust in the middle of a system upgrade :P03:00
nashSilly quesion - there isn't 41 ships there is there?03:01
* nash will have lunch shortly03:01
nashAnyway - email me the data, and I'll take a look later03:02
mithrohttp://mithro.dyndns.org/temp/error.png03:02
nashHave a lot to do today... so I'll probably look tonight03:02
mithrookay, thats fair03:02
nashHmm... looking at that... it appears to be a problem parsing the message - are you using a different server to tpserver?03:03
nashAnd it is minisec?03:03
mithroyes, and tpserver-cpp03:03
nashhmm... weird.03:03
nashI've never seen that.03:03
nashThe dat ais lsit of fleets - so it's got grabage from the server.03:03
nashMay be a parsing issue03:04
nashLeave it for now then, and I'll see if I can see the problem later03:04
mithrookay03:04
mithrobblr03:43
*** mithro has quit IRC03:43
*** jasmine has quit IRC03:51
*** jasmine has joined #tp03:52
*** mithro has joined #tp04:02
*** jasmine has quit IRC04:04
*** jasmine has joined #tp04:05
*** mithro has quit IRC04:10
*** mithro has joined #tp04:34
*** mithro has quit IRC04:37
*** mithro has joined #tp05:00
*** mithro has quit IRC05:02
*** mithro has joined #tp05:03
*** mithro has quit IRC05:07
*** jotham has joined #tp05:30
*** zipola has quit IRC05:40
*** tpb has joined #tp06:00
*** ChanServ sets mode: +o tpb06:00
*** mithro has joined #tp07:02
*** zipola has joined #tp09:49
* nash is back10:49
nashheyo mithro10:49
nashYou around?10:49
mithroam now11:04
nashCool11:04
mithromight be a bit delays, have a friend over - fixing his windows (shudder)11:04
nashSo installing ubuntu tonight then?11:04
nashWhat did you do exactly to get that problem? Was it a new game etc?11:05
mithronash: just started a new game and it was there from the beginning11:19
jothamis there any way battleviewer will be able to tell which fleet is resident and which fleet is arriving (if pertinent)11:23
jothamright now i was just imagining the planets being there then all the ships randomly warping in/jumping in11:25
jotham(at the beginning of the battle)11:25
mithrojotham: not really11:26
jothamok np11:26
nashHm... Can you try removing the file edje/basic.edj11:26
nashmithro: Otherwise I have no idea - do you get everytime?11:26
mithroyes11:26
jothamanyone here ever used (with success) py2exe to create a service?11:27
jothamwin32 service11:27
jothamer w/w11:28
TBBlemithro: pong11:30
mithroTBBle: I wanted to chat to you about packaging some time11:31
mithrospecifically media packaging11:31
mithrojotham: not a service11:32
TBBleWell, I'm here-abouts, but I'm just trying out a PS2 game...11:32
nashmithro: What architecture do you have?11:32
jothami was looking at making a service to share clipboards (via explicitly pressing f12 or something) between my ubuntu box and my windows box11:32
mithronash: i68611:32
nashNot that problem then :-/11:33
jothambut clipboarding in linux is actually freaky-deaky to the extreme, beyond my actual initial understanding of how freaky deaky it is11:33
jothamsince it's a callback at paste time, so i need to make a service/daemon for linux11:33
jothamand figure i should for win32 too11:33
nashmithro: Well for now I suggest just starting the AI up, and playin gagainst it - ignore what is in it's windpw11:33
mithroTBBle: will you be around in 30 minutes?11:34
TBBleYeah.11:36
mithrookay will chat then11:36
TBBleCool.11:36
nashmithro: But I really can't reproduce11:39
mithronash: will investigate in a bit11:39
nashI can reproduce the crash BTW... thats just a case of "I haven't implemented it yet"11:40
nashin fact:11:40
nash         /* FIXME: Handle no unread messages */11:40
nashIs already there11:40
nashback shortly11:40
*** zipola has quit IRC11:42
*** zipola has joined #tp11:42
nashmithro: Don't worry too much about finding the bug... That code needs a lot of new features, so it will be modified a lot11:54
nashMain thing I care about now is the AI side11:54
nashAnyway - I'm off to bed.11:56
nashTalk to you tomorrow11:56
*** tuna-fish has joined #tp12:03
mithroTBBle: ping?12:16
mithroTBBle: what ps2 game you trying?12:18
TBBleFree Running12:20
TBBleIt's like Tony Hawk except on foot and with the controls somewhat less usefully grouped on the buttons.12:23
TBBleAfter the training section, it got about 11 minutes out of me before I became bored of it.12:24
mithroTBBle: so, I've been rewriting the media-downloading stuff12:27
mithroi'm hoping to be able to package the media so that you only have to download updates12:28
mithroTBBle: there are also new versions up on the web (just havn't been announced yet)12:28
mithroso I wanted to check that what i've done would be okay with debian policy12:28
TBBleTo use prepackaged media with updates outside the packaging system, you'd have to have the media going to /usr/share/ somewhere, and have the updates download to a user's home directory somewhere, and have the media loader load from there by preference, maybe with a script that can remove stuff from the user's directory when it's available in the /usr/share/ area...12:31
*** jasmine has quit IRC12:31
mithroTBBle: what I was thinking was12:32
mithro/usr/share/games/tp/<media server>/<media>12:32
mithrothen if you download an update it puts it in ~/.tp/media/<media server>/<media>12:32
TBBleYeah, that looks reasonable. What's a media server in this context?12:32
mithroand if a newer one appears in /usr/share it will remove the older one from ~/.tp12:33
mithro/home/tim/.tp/media/http---localhost-media-client-/common/object-icons/planet.png12:33
mithro/home/tim/.tp/media/http---localhost-media-client-/common/object-icons/planet.png.meta12:33
mithro/home/tim/.tp/media/http---localhost-media-client-/common/object-icons/star.png12:33
mithro/home/tim/.tp/media/http---localhost-media-client-/common/object-icons/star.png.meta12:33
TBBleOh, OK.12:34
mithroas there may be multiple "packages" from different hosts12:34
mithrothe .meta files have information about the date, and md5sum, etc12:34
mithroeach server also has a media.gz which is a compress compilation of the .meta files12:35
mithrothe media.gz should probably be generated on package install (rather then distributed)12:39
mithro(As you probably want to distribute 3d media seperate to 2d media12:41
TBBleDoes it have to be gzippedd up?12:41
mithroTBBle: not necessarily12:42
mithroany reason why it would be better non-installed?12:43
mithros/installed/gzip/12:43
mithroTBBle: do you think you could to the debs for the client this weekend?12:43
TBBleI reckon I can knock them over tomorrow.12:44
mithroTBBle: that would be cool12:44
TBBleWell, if they're being distributed seperately, I don't see a need to combine them on install.12:44
mithroI can give you an account so they can upload the debs if you want12:45
TBBleJust pulled the libtpproto-{cpp,py} and libtpclient-py darcs repos... libtpproto-py appears to be missing things from its debian/ directory12:48
mithromissing?12:48
mithrotim@ultraslim:~/oss/tp/libtpproto-py$ cd debian/12:49
mithrotim@ultraslim:~/oss/tp/libtpproto-py/debian$ ls12:49
mithrochangelog  compat  control  copyright  rules  watch12:49
mithrobtw you should be able to remove this line too12:49
mithro        # This piece of magic makes the tp module seperation work12:49
mithro        cp tp/__init__.py $(CURDIR)/debian/python-tp-netlib/usr/lib/python2.4/site-packages/tp/12:49
mithroJLP discovered the "proper" way to fix that12:50
TBBletbble@su:~/code/tp/libtpproto-py$ ls debian/12:50
TBBlechangelog  control  rules  watch12:50
mithroSkipping boring file debian/compat12:51
mithroSkipping boring file debian/copyright12:51
mithro?12:51
TBBleO_O12:51
mithroshould they be boring?12:51
TBBleNo...12:51
TBBletbble@su:~/code/tp$ cat old/libtpproto/libtpproto-py-0.2.1/debian/compat12:52
TBBle512:52
TBBletbble@su:~/code/tp$ cat old/libtpproto/libtpproto-py-0.2.1/debian/copyright12:52
TBBleThis package was debianized by Paul Hampson <[email protected]> on12:52
TBBle...12:52
mithrofixing12:52
* TBBle can't imagine what darcs means by 'boring'.12:52
mithroas in computer generated files which shouldn't be in scm12:53
CIA-20[email protected] * libtpproto-py/ (5 files in 2 dirs): Fixed missing debian files.12:53
TBBleAh.12:53
mithroTBBle: give that a try12:53
TBBleHmm. Current release is 0.2.2 of the libtp*-py and pyclient-pywx, I presume next release will be 0.2.3?12:53
TBBleIs there a way to mark a file executable in darcs?12:55
mithrosadly no12:56
mithro0.2.3 is the yet to be announced release12:57
mithros/0.2.3/0.2.2/12:58
mithroeven thought it's avalilable on the website12:58
CIA-20[email protected] * web/downloads/tpclient-pywx/ (2 files): Fixed a small tab error.13:00
TBBleWhere on the website?13:00
mithrohttp://www.thousandparsec.net/tp/downloads.php#tpclient-pywx13:01
tpb<http://ln-s.net/JkJ> (at www.thousandparsec.net)13:01
mithrohttp://www.thousandparsec.net/tp/downloads/tpclient-pywx/tpclient-pywx-0.2.2.tar.bz213:01
tpb<http://ln-s.net/JkK> (at www.thousandparsec.net)13:01
TBBleOh, sorry, missed the s//13:01
TBBleOK, 'cause the darcs repos all still have 0.2.1 in the debian/changelog.13:01
TBBleOoops, and the debian/copyright in libtpproto-py is wrong, it says GPL.13:03
TBBleOh man. I'll have to go through these more carefully tomorrow.13:03
TBBleDo you want me to make them all neatly 0.2.2 and you can reroll the tarballs, or just aim for 0.2.3?13:03
mithrowell the debian stuff shouldn't be in the general release tarballs anyway?13:04
mithro(as it's debian specific?)13:05
TBBleDepends. I personally don't mind it being so.13:06
TBBleFor FreeRADIUS, I'd have a version in the tarballs as xxx-0, and the first version into the Debian archive would be -1, and I'd be putting those into a branch in the FreeRADIUS CVS tree anyway, so as to not have to maintain a seperate tree myself.13:07
TBBleI'll have to think on it a little.13:09
mithrookay13:09
mithroi'm happy to keep in darcs, but would prefer not to have in the "neutral" tarballs13:09
TBBleWell, if you'd prefer not to have it in the release tarballs, that makes that part easy at least.13:10
TBBleAlthough in that case, a seperate repo might be better, since the debian/ revisioning is then distinct from the upstream revisioning.13:11
TBBleHmm. Will have to think on it.13:11
CIA-20[email protected] * web/downloads/ (4 files in 2 dirs):13:12
CIA-20Windows binaries.:13:12
CIA-20 tpclient-pywx 0.2.213:12
CIA-20 tpserver 0.4.113:12
TBBleI guess we want to keep a darcs copy of the debian/ directory with the code, so that someone can grab and build it at any time. And a seperate set of debian/ directories to build the release .deb files against. If this was going into the Debian archive, _it_ would be the revision history for the debian/ directories, combined with something like dpatch. So I guess that's easy enough.13:13
*** tuna-fish has quit IRC13:25
*** tuna-fish has joined #tp13:25
TBBleOK, I'm sleepy and am going to bed. TTFN.13:27
*** mithro has quit IRC13:47
*** mithro has joined #tp14:31
*** mithro has quit IRC14:33
*** mithro has joined #tp15:10
* nash is back22:01
*** tuna-fish has quit IRC23:02
*** JLP has quit IRC23:34

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