Log for #openttdcoop.devzone on 9th April 2011:
Times are UTC Toggle Colours
00:01:50  *** supermop has joined #openttdcoop.devzone
00:09:27  *** KenjiE20 has quit IRC
00:39:48  *** Lakie has quit IRC
02:05:55  <Brot6> Nutracks - Revision 185:c14a72fc9fc3: Improved high speed tracks (oberhuemer) @
02:06:43  <Brot6> nutracks: update from r184 to r185 done -
02:06:50  <Brot6> nutracks: compile of r185 still failed (#2510) -
04:36:49  *** supermop has quit IRC
04:49:25  *** andythenorth has joined #openttdcoop.devzone
05:06:01  *** andythenorth has quit IRC
05:42:50  *** andythenorth has joined #openttdcoop.devzone
05:56:14  <andythenorth> morning
05:56:39  *** andythenorth has quit IRC
06:09:51  <dihedral> planetmaker, that one bug you posted is way more tricky than i expected
06:09:56  <dihedral> ^^
06:14:38  *** andythenorth has joined #openttdcoop.devzone
06:19:17  <dihedral> \o/
06:19:29  <dihedral> easier than i thought just a second ago :-D
06:24:59  *** andythenorth_ has joined #openttdcoop.devzone
06:27:59  <Brot6> Grapes - Revision 135:35beb08f4804: -Fix: (#2505) only use the URLClassLoader if we actually find... (dih) @
06:31:22  <Brot6> Grapes - Bug #2505 (Closed): errors with empty plugin dir (planetmaker) @
06:31:22  <Brot6> Grapes - Bug #2505 (Closed): errors with empty plugin dir (dih) @
06:31:59  *** andythenorth has quit IRC
06:34:23  *** Ruudjah has joined #openttdcoop.devzone
06:34:25  <Ruudjah> better
06:35:36  <Ruudjah> whats exactly the problem with current impl of plugins?
06:35:40  <dihedral> Ruudjah, little info about this channel - most messages are things like
06:35:42  <dihedral> <Brot6> [09 Apr 2011 - 04:06:50] nutracks: compile of r185 still failed (#2510) -
06:35:44  <Ruudjah> by the way, I never wrote plugin infra
06:36:28  <dihedral> well - for one thing the problem i actually was thinking of in the back of my head when saying 'plugins' was 'joan'
06:36:43  <Ruudjah> but from some perspective you can view the OpenSettlers code as one big plugin infra
06:36:55  <dihedral> you need to extend a class from joan and implement methods, rather than using some event system
06:37:01  <dihedral> which i would personally find more attractive
06:38:01  <Ruudjah> events are more rough to debug
06:38:19  <andythenorth_> morningz
06:38:55  <Ruudjah> event fired -> eventmanager calls method on an interface -> interface impl has bug and throws NPE -> IDE shows bug at calling point of event interface
06:39:47  <Ruudjah> but if you know this, it gets easier
06:40:04  <Ruudjah> Not completely grapsing what u just said, dihedral
06:41:45  <dihedral> planetmaker, Ammler, :
06:42:29  <dihedral> Ruudjah,
06:43:09  <dihedral> Ruudjah, this is the referenced OpenTTD class:
06:44:19  *** Ruudjah2 has joined #openttdcoop.devzone
06:44:35  <Ruudjah2> freeze
06:44:45  <Ruudjah2> dihedral: can you pass that link again pls?
06:45:49  <dihedral> <dihedral> Ruudjah,
06:45:49  <dihedral> <dihedral> Ruudjah, this is the referenced OpenTTD class:
06:45:49  <Ruudjah2> lol
06:45:59  <Ruudjah2> just realized I am actually eating grapes :D
06:46:02  <Brot6> HEQS "Heavy Equipment" Set - Feature #2511 (New): Forklift truck (andythenorth) @
06:46:31  <Ruudjah2> so plugins need to derive from OpenTTD
06:46:43  <dihedral> no - only what uses joan
06:46:56  <dihedral> sorry - that was my fault - i was not thinking of plugins i was thinking of joan
06:47:04  <Ruudjah2> ok
06:48:09  *** Ruudjah has quit IRC
06:48:31  <Ruudjah2> So - say I wanna make a webapp showing status of an ottd server - i;d talk to OpenTTD
06:49:21  <Ruudjah2> and instead of onWhateverSomething you want to have events, am I understanding you correct?
06:49:33  <dihedral> yarp
06:49:40  <dihedral> at least that is what i am considering :-P
06:50:33  <Ruudjah2> So I can do openTTD.addClientJoinedHandler(this)
06:50:52  <Ruudjah2> in some kind of UI class representing a list of clients
06:51:21  <Ruudjah2> the UI class obviously implementing ClientJoinEventHandler interface
06:52:08  <dihedral> i think this would be a way cleaner implementation, yes
06:52:21  * Ruudjah2 agrees
06:53:03  <Ruudjah2> e.g.
06:53:04  <Webster> Title: src/java/soc/gwtClient/game/widgetsSvg/visuals/ at master from generateui/OpenSettlers - GitHub (at
06:53:15  <Ruudjah2> line 44
06:53:46  <Ruudjah2> Next version (alpha5 of OS) I think I will be starting to use protobuf or thrift
06:53:54  <Brot6> Java OpenTTD Admin Library - Bug #2504 (Confirmed): Don't crash with wrong pw (timeout) (planetmaker) @
06:53:54  <Brot6> Java OpenTTD Admin Library - Bug #2504: Don't crash with wrong pw (timeout) (planetmaker) @
06:53:54  <Brot6> Java OpenTTD Admin Library - Bug #2504 (Confirmed): Don't crash with wrong pw (timeout) (dih) @
06:54:31  <dihedral> however i'd like to postpone that chnage (events) until v2
06:54:49  <Ruudjah2> then you'd have RobberData (simple class just for the data of a robber), which you pass to Robber when instantiating, and Robber itself you pass to RobberVisual when instantiating
06:55:16  <Ruudjah2> to complete the example
06:55:37  <Ruudjah2> How are commands defined in OpenTTD itself?>
06:55:48  <Ruudjah2> Command classes implemented I guess
06:56:08  <Ruudjah2> Somewhere I read that each command has 64 bytes each
06:56:28  <Ruudjah2> Depending on the implementation, it might even server OpenTTD to use protobuf
06:56:33  <Ruudjah2> *serve
06:56:58  <Ruudjah2> then you can share the protobuf implementation across every other thing talking to OpenTTD for the data
06:57:11  <Ruudjah2> be it js/ruby//python/java
06:57:55  <Ruudjah2> So to get started i'd build grapes, berries, joan and then there's some example simple project I suppose?
06:58:35  <Ruudjah2> Then I'd be able to add the EventBus stuff
06:59:14  <dihedral> berries are plugins - a bunch of individual projects bundled by one parent pom file
06:59:30  <dihedral> docommands are not the same as 'commands' ;-)
06:59:50  <dihedral> most of the enums (as you asked yesterday) are derived from OpenTTD enums
06:59:58  <dihedral> in order to spare confusion ;-)
07:00:56  <Ruudjah2> right
07:01:06  <dihedral> planetmaker, if no password is set when starting openttd -D, the admin port will not be opened
07:01:06  <Ruudjah2> You know protobuf
07:01:07  <Ruudjah2> ?
07:01:10  <dihedral> nope
07:01:24  <Ruudjah2> basically its a language it describe messages between apps
07:01:38  <Ruudjah2> in ottd context Commands between server and client
07:01:38  <dihedral> planetmaker, which explains the timeout - for me i get a refused ;-)
07:02:03  <Ruudjah2> You can then generate code for several langs
07:02:04  <dihedral> Ruudjah2, openttd uses it's own protocol, and it's binary ^^
07:02:27  <Ruudjah2> protobuf is a lang to define protocolls
07:02:36  <dihedral> but it already is defined
07:02:46  <Ruudjah2> in c++ code
07:03:07  <dihedral> yep - and that kind of puts down the foundation of how to communicate with the openttd server
07:03:17  <dihedral> i very highly doubt it will change
07:03:22  <Ruudjah2> If (idk, since im not into ottd code) you can move that to protobuf, you'd abstract that away (and keep the same efficiency)
07:03:49  <Ruudjah2> but maybe little too far fetched
07:03:49  <dihedral> but it would be a different style protocol ;-)
07:03:56  <Ruudjah2> not sure
07:04:08  <Ruudjah2> depending on current impl it might stay the same
07:04:30  <Ruudjah2> But for joan/beries/grapes think for now a bit too far
07:04:49  <Ruudjah2> for simple data definitions, using protobuf for joan not bad idea
07:04:58  <dihedral> it would not help, because joan has to communicate with openttd and openttd defines the protocol ^^
07:05:29  <Ruudjah2> You'd generate the code to serialize/deserialize message data for your desired language
07:05:31  <dihedral> it would probably be more work to change that, than to complete joan into a state i would love to call v1
07:05:47  <Ruudjah2> (in case for implementation of PB - protobuf - in ottd)
07:06:06  <Ruudjah2> for joan/b/g using protobuf can still be usefull
07:06:34  <Ruudjah2> less code, easier programming, etc etc
07:07:09  <Ruudjah2> simple backend plugging, no SQL neede
07:07:10  <Ruudjah2> d
07:07:35  <Ruudjah2> just throw the classes to an object database, never versioning problems b/c nature of PB
07:08:29  <Ruudjah2> Idk about performance of EventBus - I suppose you won't throw hundreds of events per sec?
07:08:54  *** ODM has joined #openttdcoop.devzone
07:09:06  <Ruudjah2> OpenSettlers only maxes about 10 events/sec, no hickup detected (even for compiled javascript in browser)
07:09:22  <Ruudjah2> matter of trial/error think
07:15:47  <Ruudjah2> What's the state you'd call it v1?
07:18:17  *** Ruudjah has joined #openttdcoop.devzone
07:18:30  <Ruudjah> I hate my conn
07:21:14  <dihedral> i'd call it v1 once everything openttd supports for that ptorocol is implemented ^^
07:21:36  <dihedral> the protocol is implemented, however not all data is reflected
07:22:52  <Ruudjah> you have a link for the 2nd protocoll for bots/management impl in ottd?
07:23:49  *** Ruudjah2 has quit IRC
07:25:47  <dihedral> <- any of the _admin stuff ;-)
07:30:51  <dihedral> ...?
07:30:56  <dihedral> enums? constants? ...
07:31:27  <dihedral> better than reading a bunch of numbers ;-)
07:31:56  <Ruudjah> thats true
07:34:34  <Ruudjah> I prefer interfaces and lowercase though ;)
07:37:09  <dihedral> that's a preference ;-)
07:37:21  <dihedral> + uppercase helps distinguish
07:37:58  <Ruudjah> that's a preference --> not always.
07:39:42  <dihedral> well, then i can argue that it is not a preference in this case either, the coding style requires it :-D
07:51:13  <Ruudjah> NETWORK_ERROR_GENERAL, // Try to use this one like never
07:53:32  * dihedral grins
07:53:47  <dihedral> Ruudjah, what's wrong with that?
07:54:04  <Brot6> Grapes - Bug #2504 (Confirmed): Don't crash with wrong pw (timeout) (planetmaker) @
07:54:05  <Brot6> Grapes - Bug #2504: Don't crash with wrong pw (timeout) (planetmaker) @
07:54:05  <Brot6> Grapes - Bug #2504 (Confirmed): Don't crash with wrong pw (timeout) (dih) @
07:54:05  <Brot6> Grapes - Bug #2504: Don't crash with wrong pw (timeout) (dih) @
08:01:36  <Brot6> Grapes - Revision 136:b4877b15fb74: -Change: separate further loading from constructor (dih) @
08:01:36  <Brot6> Grapes - Revision 137:5fab8d38e7d4: -Fix: (#2504) Handle connection issues more graceful (dih) @
08:02:09  <dihedral> \o/
08:02:48  <Brot6> Grapes - Bug #2504 (Closed): Don't crash with wrong pw (timeout) (planetmaker) @
08:02:48  <Brot6> Grapes - Bug #2504 (Closed): Don't crash with wrong pw (timeout) (dih) @
08:05:54  <Ruudjah> :)
08:08:45  <Brot6> Grapes - Revision 138:fd034c75c0a1: -Fix: Handle unknown host issues, too (dih) @
08:12:28  <dihedral> that's all for today
08:12:35  <dihedral> have a nice weekend guys
08:36:39  <Ruudjah> cya
08:39:23  *** andythenorth has joined #openttdcoop.devzone
08:46:50  *** andythenorth_ has quit IRC
09:09:53  <andythenorth> ho
09:10:06  <andythenorth> interesting reaction from SAC :P
09:39:19  <Ammler> andythenorth: you once posted a error on devzone, is it fixed? :-)
09:39:57  <andythenorth>
09:40:04  <andythenorth> 500 if you try to view something
09:40:11  <andythenorth> might be limited to FIRS repo, not sure
09:42:18  <Ammler> it isn't
09:42:27  <Ammler> but it happens only logged in
09:42:32  <andythenorth> how interesting
09:42:48  <Ammler> I didn't test that on the stage server :-)
09:44:34  <Ammler> found issue: plugin review
10:03:02  *** andythenorth has left #openttdcoop.devzone
10:14:08  *** frosch123 has joined #openttdcoop.devzone
10:58:07  *** KenjiE20 has joined #openttdcoop.devzone
11:11:03  *** KenjiE20 has quit IRC
11:19:45  *** KenjiE20 has joined #openttdcoop.devzone
11:22:05  <Brot6> [releases] openSUSE API not reachable, sleep an hour and try again...
11:22:05  <Brot6> [nightlies] openSUSE API not reachable, sleep an hour and try again...
12:15:02  *** DanMacK has joined #openttdcoop.devzone
12:45:14  *** KenjiE20 has quit IRC
12:53:16  *** KenjiE20 has joined #openttdcoop.devzone
13:22:05  <Brot6> [releases] openSUSE API not reachable, sleep an hour and try again...
13:22:05  <Brot6> [nightlies] openSUSE API not reachable, sleep an hour and try again...
14:52:25  *** andythenorth has joined #openttdcoop.devzone
15:01:24  <Brot6> HEQS "Heavy Equipment" Set - Feature #2512 (New): Reduce lifetime of trams (andythenorth) @
15:41:27  *** ODM has quit IRC
15:58:48  *** Lakie has joined #openttdcoop.devzone
17:18:13  <Brot6> nutracks: update from r179 to r185 done -
17:18:20  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r750), 32bpp-extra (r39), ai-admiralai (r75), ai-aroai (r25), ailib-common (r21), ailib-direction (r17), ailib-list (r32), ailib-string (r29), ailib-tile (r16), airportsplus (r73), basecosts (r25), belarusiantowns (r8), bros (r52), chips (r128), comic-houses (r71), firs (r1915), fish (r617), frenchtowns (r6), german-townnames (ERROR r24), grfcodec (r828), heqs (r604),
17:18:20  <Brot6> indonesiantowns (r41), manindu (r7), metrotrackset (r56), narvs (r37), newgrf_makefile (r266), nml (ERROR r1307), ogfx-industries (r12), ogfx-landscape (r58), ogfx-rv (r80), ogfx-trains (r237), ogfx-trees (r42), opengfx (r637), openmsx (r97), opensfx (r97), smts (r19), snowlinemod (r49), spanishtowns (r10), swedishrails (r198), swisstowns (r22), transrapidtrackset (r15), ttdviewer (r26), ttrs (r36), worldairlinersset (r671)
17:27:24  *** ODM has joined #openttdcoop.devzone
18:29:39  <Brot6> FIRS Industry Replacement Set - Revision 1916:dda09601d74f: Change: improve appearance of Furnitu... (andythenorth) @
18:55:01  *** DanMacK has quit IRC
19:31:52  <Brot6> FIRS Industry Replacement Set - Revision 1917:4b0d89790fc0: Change: improve appearance of Windmill (andythenorth) @
19:47:21  *** andythenorth has quit IRC
21:21:33  *** Lakie has quit IRC
21:51:09  *** ODM has quit IRC
22:59:31  *** frosch123 has quit IRC
23:47:30  *** KenjiE20 has quit IRC

Powered by YARRSTE version: svn-trunk