--- Log opened Wed May 27 00:00:02 2009
20090527 00:05:44-!- Elvish_Pillager [n=eli@68-116-175-221.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090527 00:05:49-!- Elvish_Pillage2 [n=eli@68-116-175-221.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090527 00:09:13-!- wesbot changed the topic of #wesnoth-dev to: 60 bugs, 241 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090527 00:19:45-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit ["Quitte"]
20090527 00:29:40-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090527 00:33:45 * loonycyborg needs advice
20090527 00:34:11< loonycyborg> Should I commit revert of suokko's commit or wait for someone to perhaps fix it?
20090527 00:36:02< Soliton> if that someone is you then wait i'd say.
20090527 00:37:18-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit []
20090527 00:37:46 * loonycyborg would rather not to try to fix AI stuff and leave it for someone more qualified, e.g. Crab :P
20090527 00:38:19-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090527 00:38:32< Soliton> is it really AI related? i'd guess some event just doesn't fire right.
20090527 00:39:05-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit ["ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]"]
20090527 00:46:24< noy> loonycyborg: does the commit have any redeeming value?
20090527 00:46:33-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090527 00:46:52< loonycyborg> noy: It should fix this bug: https://gna.org/bugs/index.php?10603
20090527 00:48:14< noy> ... I've got no way to assess it. 
20090527 00:48:34< loonycyborg> Perhaps suokko just tested it for the affected scenario..
20090527 00:51:00< noy> loonycyborg: make a bug report perhaps for someone to examine it?
20090527 00:51:48< noy> to see whether its worthwhile to keep it or not?
20090527 00:55:34< loonycyborg> I can add a comment to that bug report.
20090527 00:56:39-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has quit [Connection reset by peer]
20090527 01:03:21< stikonas> esr: I believe that https://gna.org/bugs/?13086 was fixed in 1.7.0, can you close it?
20090527 01:04:05-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090527 01:04:39< stikonas> i can not reproduce it any longer
20090527 01:05:03< esr> OK, I'll close.
20090527 01:05:52< esr> Done
20090527 01:06:00< stikonas> thanks
20090527 01:09:03-!- Doppp [n=aassdf@c-67-171-96-240.hsd1.pa.comcast.net] has quit [Read error: 104 (Connection reset by peer)]
20090527 01:09:44-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20090527 01:11:23-!- noy [n=Noy@wesnoth/developer/noy] has quit [Connection timed out]
20090527 01:13:49-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 01:15:24-!- Doppp [n=aassdf@c-67-171-96-240.hsd1.pa.comcast.net] has joined #wesnoth-dev
20090527 01:33:40-!- Smar [n=smar@freenet/translator/finnish/Smar] has quit [Read error: 110 (Connection timed out)]
20090527 01:33:50-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090527 01:42:41< Soliton> loonycyborg: looking at the AI units.. they just don't get their moves back.
20090527 01:43:01< Soliton> so the turn events don't fire for the AI it seems.
20090527 01:50:40-!- Smar [n=smar@a88-113-60-192.elisa-laajakaista.fi] has joined #wesnoth-dev
20090527 02:00:05< loonycyborg> It's weird that it doesn't affect human players then. I can play with myself with two clients on local wesnothd :P
20090527 02:00:34< loonycyborg> That is that bug doesn't happen if I do so.
20090527 02:03:37< loonycyborg> And if an AI replaces a player who had quit that doesn't seem to happen either.
20090527 02:05:05< Soliton> looks like i got a fix.
20090527 02:05:39< Soliton> is_local() does not take plain AI sides into account.
20090527 02:06:26< Soliton> not sure if anything requires that assumption though...
20090527 02:12:37< Soliton> well, seems safe to me.
20090527 02:14:07-!- LinkPie [n=Gamefrea@67.239.184.65] has joined #wesnoth-dev
20090527 02:17:16-!- BenUrban [n=benurban@68.55.19.224] has joined #wesnoth-dev
20090527 02:18:36< CIA-77> soliton * r35905 /branches/1.6/data/tools/ (unit_tree/helpers.py wesnoth/wmlparser.py): 
20090527 02:18:36< CIA-77> silence some verbose output
20090527 02:18:36< CIA-77> the verbose setting of the WML parser could be turned into an int
20090527 02:18:36< CIA-77> for better granularity if wanted
20090527 02:19:22< CIA-77> soliton * r35906 /trunk/src/team.hpp: plain AI sides are local, too
20090527 02:19:27< CIA-77> soliton * r35907 /trunk/data/tools/ (unit_tree/helpers.py wesnoth/wmlparser.py): 
20090527 02:19:27< CIA-77> silence some verbose output
20090527 02:19:27< CIA-77> the verbose setting of the WML parser could be turned into an int
20090527 02:19:27< CIA-77> for better granularity if wanted
20090527 02:21:30< Soliton> loonycyborg: if you could test if suokko's fix actually fixed that bug report that'd be cool.
20090527 02:25:31-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090527 02:27:40< loonycyborg> Soliton: How that test scenario should be installed? It doesn't look like a campaign..
20090527 02:29:45< Soliton> put it into data/add-ons? probably have to run wmllint on it first.
20090527 02:30:36-!- Polarina [n=polarina@wesnoth/translator/Polarina] has quit ["Leaving."]
20090527 02:35:35< Turuk> This just in.  RNG affected by team color.
20090527 02:36:12< Soliton> so, what's the lucky color?
20090527 02:36:33< loonycyborg> Caught AttributeError when trying to remove blueblaze_wesboard from dict <- wmllint
20090527 02:37:09-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090527 02:37:44< Soliton> turn spellcheck or whatever off?
20090527 02:38:57< Turuk> Apparently Black, but that's just for Dwarvish Thunderers
20090527 02:39:08< Turuk> Each faction responds well to a certain color
20090527 02:39:19< Soliton> ah, of course.
20090527 02:40:11< Turuk> I am more worried that he truly believes it.  The human race is doomed if this is the future.
20090527 02:40:54< Soliton> sounds like there is a colorful future ahead to me.
20090527 02:41:37-!- shikadibot [n=elynia@wesnoth/bot/shikadibot] has quit [Remote closed the connection]
20090527 02:41:37-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Remote closed the connection]
20090527 02:42:22< Turuk> Ha, very true.
20090527 02:43:34-!- Elvish_Pillage2 [n=eli@68-116-175-221.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090527 03:03:07 * loonycyborg decides to continue wrestling with wmllint tomorrow.
20090527 03:03:12-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090527 03:03:50-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090527 03:05:26-!- LinkPie [n=Gamefrea@67.239.184.65] has quit [Read error: 104 (Connection reset by peer)]
20090527 03:18:28-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has joined #wesnoth-dev
20090527 03:18:46< Sapient> I'm back
20090527 03:19:25< Sapient> and much tanner now ;)
20090527 03:37:49< esr> Sapient: Is dialogs.cpp your code?
20090527 03:38:05< Sapient> No, but I've tweaked parts of it
20090527 03:38:11< Sapient> As I have most of the files
20090527 03:39:03< Sapient> In particular, there was a time when I refactored the GUI from function-based to object-based
20090527 03:39:28< Sapient> and made it extensible
20090527 03:39:55< esr> Hmmm...I'm looking for the author oif the load_game dialog, as my attempts to refactor it into a class so I can add a button and a callback have twice failed - code looks OK but results segfault.
20090527 03:40:31< Sapient> ah, yes... I once sought to do the same change
20090527 03:40:44< Sapient> I refactored the editor's load map dialog into a class
20090527 03:41:10< Sapient> but, it wasn't as re-usable as I hoped for the load game dialog
20090527 03:41:25< Sapient> mainly because of the byzantine code of the load_game dialog
20090527 03:41:35 * esr winces
20090527 03:42:52< Sapient> are you using the construct_dialog.hpp toolkit?
20090527 03:43:12< Sapient> if so, I can assist
20090527 03:43:38< Sapient> I you're converting it to the Mordante's GUI2 system, I don't know anything about that
20090527 03:44:20< Sapient> *err, "If"
20090527 03:44:34< esr> No, I just wanted to refactor it into a clkass so I coulsd add a "Show only tip saves" toggle.  You know about my save-threading proposal, right?
20090527 03:45:02< Sapient> I scanned over it pretty fast, yeag
20090527 03:46:26< Sapient> which does you class inherit from?
20090527 03:46:57< esr> Well, the infrastructure's all done; trunk is already generating what are in effect parent links into each savefiile, and they're being read, and I have code to correctly invert the pointer chain.
20090527 03:47:14< esr> My attempts didn't have a parent class.
20090527 03:47:43< Sapient> maybe you don't understand my question about which gui system you are using...
20090527 03:48:44< esr> I was just trying to refactor the existing GUI1 code.
20090527 03:50:20< Sapient> if you call gui::dialog then that's gui1.5 (sapient) gui::show_dialog() is gui1.0 (dave)
20090527 03:50:28< esr> My attempt at a load_game class was basically the guts of the existing load function with its bodty as the initializer.  No actual inheritance frm the GUI.
20090527 03:50:42< esr> Oh, then it's gui1.5.
20090527 03:50:54< Sapient> ok, cool
20090527 03:52:11< Sapient> gui2::tdialog is gui2 (mordante)
20090527 03:52:25< Sapient> feel free to pastebin me if you have a question
20090527 04:09:48-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit []
20090527 04:15:25-!- fabi [n=fabi@f051079246.adsl.alicedsl.de] has joined #wesnoth-dev
20090527 04:16:18-!- fendrin [n=fabi@wesnoth/developer/fendrin] has quit [Read error: 60 (Operation timed out)]
20090527 04:22:41-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has quit [Nick collision from services.]
20090527 04:22:47-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has joined #wesnoth-dev
20090527 04:26:09-!- crimson_penguin [n=ben@64.201.60.216] has joined #wesnoth-dev
20090527 04:38:59-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 05:08:31-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 110 (Connection timed out)]
20090527 05:20:48-!- [Relic] [n=[Relic]@adsl-76-230-212-46.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20090527 05:21:37< [Relic]> Hello :)
20090527 05:42:43-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20090527 05:51:44< Sapient> Soliton: there?
20090527 05:52:15< Sapient> I think r35896 is not a good idea...
20090527 05:52:39< Sapient> we would not like the user to "press any key"
20090527 05:52:51< Sapient> for example, Escape key
20090527 05:55:07< Sapient> I think that it is unecessary to explain all the keys that could or could not be pressed and better to just say spacebar for simplicity's sake
20090527 05:56:34< Sapient> if they need to learn about skipping messages with the Escape key, that could be done in scenario 2 perhaps
20090527 06:06:29< CIA-77> sapient * r35908 /trunk/data/campaigns/tutorial/scenarios/1_Tutorial.cfg: stick with the (safer) spacebar suggestion
20090527 06:09:17-!- wesbot changed the topic of #wesnoth-dev to: 59 bugs, 241 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090527 06:19:02-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Remote closed the connection]
20090527 06:22:23-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090527 06:33:05-!- BenUrban_ [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev
20090527 06:52:31-!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Read error: 110 (Connection timed out)]
20090527 06:52:55-!- BenUrban_ is now known as BenUrban
20090527 06:59:27-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090527 07:00:01-!- Sapient [n=sapien-x@wesnoth/developer/sapient] has quit []
20090527 07:01:36-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 07:07:06-!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev
20090527 07:11:41-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Success]
20090527 07:13:41-!- [Relic] [n=[Relic]@adsl-76-230-212-46.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20090527 07:27:08-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090527 07:27:23-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 07:38:19-!- Sirp [n=me@wesnoth/developer/dave] has quit ["leaving"]
20090527 08:03:33-!- silene1 [n=plouf@ASte-Genev-Bois-152-1-21-116.w83-114.abo.wanadoo.fr] has joined #wesnoth-dev
20090527 08:04:48-!- silene [n=plouf@wesnoth/developer/silene] has quit [Read error: 110 (Connection timed out)]
20090527 08:07:01-!- DDR [n=chatzill@66.183.125.196] has quit ["ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]"]
20090527 08:07:27-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090527 08:16:55-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090527 08:17:10-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 08:24:24-!- maxy [n=maxy@80-219-0-199.dclient.hispeed.ch] has joined #wesnoth-dev
20090527 08:32:15-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090527 08:32:17< euschn> hi
20090527 08:41:55< euschn> YogiHH: I updated the uml model: http://stud3.tuwien.ac.at/~e0204097/wesnoth/uml/ . I added play_controller, a new class tod_manager, a ToConfig interface and small changes to replay and controller_base
20090527 08:42:01-!- stikonas [n=and@213.164.112.217] has joined #wesnoth-dev
20090527 08:43:34 * YogiHH checks
20090527 08:44:50-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090527 08:50:08< YogiHH> euschn: did do_replay get a play_controller parameter passed before or did you add that to the call?
20090527 08:50:22< euschn> yes
20090527 08:50:53< YogiHH> yes before or yes added? ;-)
20090527 08:51:06< euschn> err yes before :)
20090527 08:51:10< euschn> sorry just got up haha
20090527 08:51:19< YogiHH> np :-)
20090527 08:52:02< euschn> i on made the changes i noted in the diagram, that is, removing gamestatus and teams arguments
20090527 08:52:07< euschn> s/on/sonly
20090527 08:53:09< YogiHH> euschn: ok, what do we need game_state.sides_member for? I ask because it is kind of redundant, as it will be contained in either snapshot or starting_pos.
20090527 08:57:33< euschn> YogiHH: good point, I added it as kindof replacement for the teams vector, but of course it can be added to snapshot
20090527 09:12:55< YogiHH> euschn: i am not sure if we need it at all. If for example the teams are built and we need to pass the side config, we can evaluate it on the fly with a local collection.
20090527 09:14:16< euschn> YogiHH: yes thats what I ment, youre right I will remove sides_
20090527 09:16:30< YogiHH> euschn: What about game_state.get_teams()/get_team_info(), can we move that as well (for example to play_controller)?
20090527 09:18:20< YogiHH> euschn: IIRC, there is also a class team_manager, you might want to check that out if you did not do that yet. Might be a good home for those functions, too, as well as get_first_human_team()
20090527 09:18:29< euschn> YogiHH: again, I agree, this is one of the points Im still thinking about. get_team_info() basically populates highlevel objects (belonging to play_controller) from a config (belonging to game_state) so i guess either is fine
20090527 09:19:14< YogiHH> oh, i wasn't aware it populates them, too. I thought it would be kind of a getter.
20090527 09:20:15< YogiHH> euschn: Can you check if constructors are a possible alternative, then?
20090527 09:20:16< euschn> YogiHH: it definitely fills up the teams list using the config
20090527 09:21:23< euschn> YogiHH: they might be. currently, get_human_info is currently called in play_controller::init, which is sort of a constructor
20090527 09:22:10< euschn> as for teams_manager: I was aware of it, it just does not do a lot atm. might be a good idea to add get_first_human_theam there
20090527 09:22:11< YogiHH> euschn: two more minor things, i'd rename "ToConfig" to "SavegameConfig" (just the class, not the method)
20090527 09:22:27< euschn> sure, it was more like a placeholder name anyways
20090527 09:23:45< Soliton> Sapient: well, what other key is special besides escape?
20090527 09:24:59< YogiHH> euschn: and i'd consider to move all those separate primitives in game_state to a container object, like for example "game_classification". I was thinking about game_info, but there is this golden rule not to name your classes "xxxInfo". I don't have a better idea atm, though.
20090527 09:26:09< euschn> YogiHH: ok, will do that - and I will keep that golden rule in mind :)
20090527 09:26:23< Soliton> Sapient: not sure why i thought any key works though since it looks like only space and enter work...
20090527 09:26:31< YogiHH> euschn: can you imagine why i'd like to have a separate object for it?
20090527 09:27:13< euschn> YogiHH: well it gives mroe structure and better overview to game_state
20090527 09:27:27< YogiHH> that's one thing, yes
20090527 09:27:28< euschn> considering youre question, I guess you have another reason too
20090527 09:27:43< YogiHH> another one is using the SavegameConfig interface
20090527 09:28:07< euschn> i see - as most of the primitves are meta/toplevel info
20090527 09:29:13< YogiHH> if every savegame-relevant object implements the interface, we can iterate over an array of SavegameConfig to create the snapshot (or the whole savegame config, for that matter)
20090527 09:32:05< euschn> YogiHH: I see, good idea. Ill update the model accordingly
20090527 09:37:24< CIA-77> silene * r35909 /trunk/src/team.hpp: Removed dead code: a human AI (whatever that means) is an AI.
20090527 09:39:57< Ivanovic> moin
20090527 09:43:27-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090527 09:47:33-!- silene1 [n=plouf@ASte-Genev-Bois-152-1-21-116.w83-114.abo.wanadoo.fr] has quit ["Leaving."]
20090527 09:47:56-!- fabi [n=fabi@wesnoth/developer/fendrin] has quit [Remote closed the connection]
20090527 09:49:01< CIA-77> ivanovic * r35910 /trunk/ (18 files in 17 dirs): updated Czech translation
20090527 09:49:06< CIA-77> ivanovic * r35911 /branches/1.6/ (12 files in 11 dirs): updated Czech translation
20090527 09:51:16-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090527 09:56:57-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 10:03:20-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090527 10:11:05-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Remote closed the connection]
20090527 10:11:33-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090527 10:26:13-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090527 10:28:16-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090527 10:39:47-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090527 10:48:55-!- Polarina [n=polarina@wesnoth/translator/Polarina] has joined #wesnoth-dev
20090527 10:55:00-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has joined #wesnoth-dev
20090527 10:58:30< euschn> YogiHH: I had a look at teams_manager. I noticed that it does not have a reference to a vector<team> of its own, but uses a static vector of teams, defined at team.cpp:35
20090527 10:59:01< euschn> when moving get_first_human_team to teams_manager, can we guarantee this static list of teams is always the same as the one of play_contrioller?
20090527 10:59:52< euschn> they are set to the same reference in the play_constroller constructor, but some other class might access the static vector?
20090527 11:00:07< euschn> s/access/modify
20090527 11:02:06-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090527 11:02:33< Ivanovic> off to uni, cui
20090527 11:09:01< YogiHH> euschn: I'd be very surprised. IIRC, the team manager is set up within the initialization of play_controller, so there shouldn't be any additional manipulation from outside.
20090527 11:12:34< euschn> YogiHH: ok
20090527 11:23:47-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Remote closed the connection]
20090527 11:39:55-!- Crab_ [n=Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20090527 11:42:18-!- loonybot [n=loonybot@79.139.246.189] has joined #wesnoth-dev
20090527 11:43:16-!- loonycyborg [n=sergey@79.139.246.189] has joined #wesnoth-dev
20090527 11:44:41-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has joined #wesnoth-dev
20090527 11:44:58< cristi_ceata> hello! can anybody tell me a Perl interpreter for linux?
20090527 11:48:35 * loonycyborg fails to parse
20090527 11:50:12-!- cristi_ceata [n=morpheus@p22.eregie.pub.ro] has left #wesnoth-dev []
20090527 11:51:38 * Rhonda looks flabbergasted
20090527 12:42:29-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090527 12:42:44-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 13:07:22-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 13:15:34-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090527 13:22:47-!- Elvish_Pillager [n=eli@68-116-173-131.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090527 13:52:55-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has quit [Success]
20090527 14:02:57-!- Elvish_Pillage2 [n=eli@68-116-173-131.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090527 14:04:06-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090527 14:04:35-!- shadowmaster [n=ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20090527 14:04:43-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090527 14:05:00-!- Elvish_Pillager [n=eli@68-116-173-131.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090527 14:06:06-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20090527 14:06:32-!- shikadibot [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 14:07:35-!- elynia [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 14:16:53-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090527 14:17:08-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 14:55:26-!- elynia [n=elynia@wesnoth/bot/shikadibot] has quit ["Lost terminal"]
20090527 14:59:11-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090527 14:59:18< euschn> hi again
20090527 15:10:34-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090527 15:21:16-!- ettin_ [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090527 15:26:04-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has left #wesnoth-dev []
20090527 15:34:33-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 110 (Connection timed out)]
20090527 16:07:06-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 16:18:40-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090527 16:21:13-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has quit [Remote closed the connection]
20090527 16:22:41-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090527 16:24:01< boucman> hey all
20090527 16:24:17< Crab_> hi boucman
20090527 16:24:53< boucman> Crab_: I arrived home 10' ago, so I won't comment on your commits right away :P
20090527 16:25:08< boucman> did you need to discuss something in particular ?
20090527 16:26:26< Crab_> boucman: there's only one 'my commit' (last week, that 1st part with various contexts). Yes, I've got a question, but it can wait. ping me somewhen when you've free.
20090527 16:28:05< boucman> k
20090527 16:36:29-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090527 16:38:00< boucman> Crab_: ok, i'm here
20090527 16:38:33< boucman> (havn't read your commit yet, though)
20090527 16:41:42< Crab_> the question is: I have a number of different AI contexts (I will document them after I finish this part of my work). Their purpose is: 'side_context' holds only side number, 'readonly_context' holds functions which are 'answers to questions about the game state',
20090527 16:41:43< Crab_>   'readwrite_context' holds functions which modify the state of the game, 'default_ai_context' will hold the functions which are used for default ai decisions (then come the various AIs and AI parts)...
20090527 16:42:18< boucman> ok
20090527 16:42:19< Crab_> there's need only  for 1 copy of side_context. there's need only for 1 copy of readonly and readwrite contexts (they contain no decisions)
20090527 16:42:36< Crab_> default_ai_context contains a lot of decisions.
20090527 16:42:59< Crab_> so, it makes sense to make two copies of default ai context - 'stable' and 'dev'
20090527 16:43:16< Crab_> to allow to change that functions and then test the outcome
20090527 16:43:21< boucman> k
20090527 16:43:29< Crab_> there's two ways to make those two copies
20090527 16:43:55< Crab_> 1) on a context-level granularity : two implementations of that default_ai_context interface
20090527 16:44:36< Crab_> 2) on a function level granularity : one implementation of  default_ai_context interface that delegates some functions to some other objects
20090527 16:44:44< Crab_> so, what's better ?
20090527 16:45:09-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit ["ChatZilla 0.9.84 [Iceweasel 3.0.6/2009050519]"]
20090527 16:45:30< boucman> hmm
20090527 16:45:57< boucman> the point is to allow other AI to reuse the decision proces of the default AI
20090527 16:46:10< boucman> I don't think such users should now which AI they are using underneath
20090527 16:46:30< boucman> since they shouldn't rely on any particular behaviour of the dev AI...
20090527 16:46:44< Crab_> boucman: the point "allow other AI to reuse the decision proces of the default AI" is met in both cases.
20090527 16:46:47-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20090527 16:46:56< Crab_> boucman: let me prepare a small pastebin snippet to illustrate the question
20090527 16:47:00< boucman> maybe we should only export the stable one, no reuse of the dev until it's changes have migrated instable
20090527 16:47:22< boucman> Crab_: yes, I've got that
20090527 16:47:40< boucman> I'm trying to "bullet list" the positive and negative aspect of both implementations
20090527 16:48:18< boucman> so far, the difference I see is that in one case the user decides if he want dev or stable, in the other it's you (as AI maintainer) which choose which one is behind the interface
20090527 16:49:12< Crab_> well, *this* is not a difference - we can publish a list "which functions are behind the interface" and name it "stable"
20090527 16:49:19-!- busfahrer is now known as buzZzZzfahrer
20090527 16:49:45< boucman> not sure what you mean...
20090527 16:52:53< Crab_> boucman: http://wesnoth.pastebin.com/m3d12a1dd
20090527 16:53:30< Crab_> 1) context-level granularity : two implementations of that default_ai_context interface
20090527 16:53:35< Crab_>  2) function level granularity : one implementation of  default_ai_context interface that delegates some functions to some other objects
20090527 16:53:53< boucman> k
20090527 16:54:16< boucman> instinctively I'd go fro the first way...
20090527 16:54:32< boucman> I'm not convinced we need to have a per function definition in WML for the default AI
20090527 16:55:02< boucman> otoh, it means that the idea of "stable" and "dev" is hardwired in the code..
20090527 16:55:28< Crab_> yes - two files, one for stable, one for dev
20090527 16:56:44< Crab_> but this is np, since our main problem is 'we must be able to tinker with the code without risking silent breakage'
20090527 16:56:58< Crab_> and two implementations is enough if we keep one of them 'stable'
20090527 16:56:58< boucman> true
20090527 16:57:22< Crab_> and then we'll be able to compare 'whatever we got at the moment' to it
20090527 16:58:04< boucman> yes as far as user (WML) is concerned, I' think way1 is enough, we don't need the fine-grained definition of way2
20090527 16:58:15< boucman> now from an internal design, I don't really mind
20090527 16:58:38< boucman> I guess doing way2 internally and not exposing it to WML doesn't make much sense, so let's go for Way1
20090527 16:59:18< Crab_> boucman: note users  and scenario creators (unless they are ai designers), will only have to specify ai identifier in both cases
20090527 16:59:50< Crab_> unless they want to override something, that is
20090527 16:59:55< Crab_> but way1 is simpler
20090527 17:00:02< boucman> ok
20090527 17:00:36< Crab_> my present stance is 'go way1 and then, if it will be needed, switch to way2'
20090527 17:01:03< boucman> Crab_: makes sense, we don't really know our need at this point, and it would be nice to move on now
20090527 17:02:48< Crab_> and also note that my current design (I'll commit part2 tomorrow in the evening - ai_composite and rca loop) allows to implement python/lua/whatever support in a simple enough way.
20090527 17:03:07< boucman> sounds good
20090527 17:05:28< Crab_> for example, to write candidate action in X, someone will need to extend a special class called 'engine' which can do the following thing: 'take a config snippet and produce an candidate_action object from it'. to write a turn stage in X,  someone will need to extend a special class called 'engine' which can do the following thing: 'take a config snippet and produce an 'stage' object from it'. and so on.
20090527 17:05:56< Crab_> this will allow to intermix c++/formula/etc candidate actions and stages
20090527 17:06:19< boucman> sounds easy enough
20090527 17:06:32< boucman> having a clear interface was the hard part, but you're almost done afaict
20090527 17:06:57< Crab_> boucman: yes. still got a few places to polish, but so far it looks good to  me.
20090527 17:08:13< Crab_> and, another question
20090527 17:08:23< boucman> sure
20090527 17:11:01-!- Sirp [n=me@wesnoth/developer/dave] has joined #wesnoth-dev
20090527 17:11:33< Crab_> boucman: I've moved to another apartment last week, and my backup computer (with ai test scripts) on it finally stands in a place where it can be 'always on' without disturbing anyone. so, I intend to finally put down a 'fire and forget' script on it which will do 'git svn rebase / compile / start testing / ... /  repeat'. what'll be a suitable schedule ?
20090527 17:12:05< boucman> you mean for refreshing your build ?
20090527 17:12:11< Crab_> yes
20090527 17:12:19< boucman> doesn't really matter, I guess
20090527 17:12:32< boucman> as long as you log your svn rev (which I know you do)
20090527 17:13:19< boucman> i'd say refresh every 2000 runs or something like that
20090527 17:13:35< Crab_> ok
20090527 17:13:37< boucman> (and having a way to have it do 2000runs of a specified run would be nice tooà
20090527 17:13:46< boucman> a specified rev, Imean
20090527 17:14:09< Crab_> there's a catch there
20090527 17:14:19< Crab_> my script, too, changes from time to time
20090527 17:14:28< boucman> :P
20090527 17:14:40< boucman> could it change the results ?
20090527 17:14:43< Crab_> so, for example, it can have different overhead in different revisions
20090527 17:14:55< Crab_> yes. for example, my next version will have per-turn logging
20090527 17:15:25< Crab_> but it hope it will be backward-compatible :)
20090527 17:16:07< Crab_> so, to compare old-to-new revision, it is needed to 'grab a new version of a script, and then use it on a old rev'
20090527 17:16:46< boucman> Crab_: well, the way it's going, you could more or less erase the database when upgrading the script (assuming it doesn't happen too often)
20090527 17:17:14< Crab_> boucman: it's better to log script version separately, imo 
20090527 17:17:16< boucman> I don't think we will need old rev more than a month old when looking for historical results, and we can alwys rerun old revs
20090527 17:17:27-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit ["The man who smiles when things go wrong has thought of someone to blame it on."]
20090527 17:17:39< Crab_> boucman: ok. this will happen after I convert a copy of default ai to rca.
20090527 17:17:59< boucman> Crab_: the problem is how to fill "missing info in older versions" but that can be done smartly
20090527 17:18:02-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 17:20:03< Crab_> boucman: well, I've extracted my 'info gathering' to one source file (ai/testing.cpp), so, if a serious need arises, I should be able to patch it to fill new info in an old version, without disturbing much of the source.
20090527 17:20:46< boucman> k
20090527 17:21:58-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has left #wesnoth-dev []
20090527 17:29:13< Crab_> boucman: and, I've rechecked my SoC must-do list for milestone 1. so far, it looks like I'm on schedule, but there's two points at the end which will require splitting formula ai into several files, plus many small changes to it. so, is there any news about what Dragonking is doing ?
20090527 17:29:19< Crab_> wesbot: seen Dragonking
20090527 17:29:19< wesbot> Crab_: The person with the nick Dragonking last spoke 16d 21h ago.  10d 14h ago was here and on the channels #wesnoth and #wesnoth-umc-dev with the message:
20090527 17:29:44< boucman> Crab_: I was away this week, so I don't know
20090527 17:29:51-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090527 17:29:51< boucman> Sirp: around ?
20090527 17:31:27-!- Noyga [n=lame-z@AVelizy-151-1-70-121.w81-249.abo.wanadoo.fr] has joined #wesnoth-dev
20090527 17:36:49< Crab_> boucman: so, if he doesn't appear in a week or two, it's possible that he'll have some merging problems.
20090527 17:39:40< boucman> Crab_: don't worry
20090527 17:39:47< Crab_> ok
20090527 17:40:01< boucman> we'll have news before that, and even if we don't, he'll have merging problems, not you ;)
20090527 17:40:38-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090527 17:41:55< Crab_> it doesn't matter 'who'll have those problems'   merging is bad, especially if it can be avoided :)
20090527 17:42:07< Crab_> but, ok  :)
20090527 17:48:42-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090527 18:02:06-!- Elvish_Pillager [n=eli@68.116.173.131] has joined #wesnoth-dev
20090527 18:02:48-!- Elvish_Pillage2 [n=eli@68-116-173-131.dhcp.oxfr.ma.charter.com] has quit [Read error: 60 (Operation timed out)]
20090527 18:19:13-!- buzZzZzfahrer is now known as busfahrer
20090527 18:24:44< Sirp> boucman: yeah what's up?
20090527 18:25:53< boucman> Sirp: Crab_ needs some info about DK's work to synchronize with him, so I wondered if you could answer a couple of questions for him  since he seem not to be able to catch DK
20090527 18:26:48< Sirp> boucman: well what's the question. I _might_ know the answer. :)
20090527 18:26:57< boucman> Crab_: :P
20090527 18:29:32< Crab_> Sirp: In around 10 days, I will begin a refactoring from src/ai/formula_ai.cpp, which will touch that file a lot. if DK has a lot of uncommitted work touching that file, it would be nice to know the 'expected date of commit'
20090527 18:30:26< Crab_> or, if his work doesn't depend on src/ai/formula_ai.cpp or, if he hasn't got uncommitted changes to that file, that's even better.
20090527 18:30:48< Sirp> Crab_: afaik DK hasn't been doing a lot of work so far due to other commitments. So I think that'll be fine.
20090527 18:31:01< Crab_> Sirp: ok, thanks.
20090527 18:36:35-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 110 (Connection timed out)]
20090527 18:43:22-!- Psyche^ [n=Psyche@f054003124.adsl.alicedsl.de] has joined #wesnoth-dev
20090527 18:54:29-!- boucman [n=rosen@wesnoth/developer/boucman] has quit ["Leaving."]
20090527 18:54:47-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090527 18:57:18-!- elynia [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 18:59:13-!- Patterner [n=Psyche@g224106109.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)]
20090527 18:59:13-!- Psyche^ is now known as Patterner
20090527 19:22:17-!- elynia [n=elynia@wesnoth/bot/shikadibot] has quit ["bye"]
20090527 19:23:58-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has quit [Connection timed out]
20090527 19:25:09-!- mordante [n=mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20090527 19:25:24< mordante> servus
20090527 19:26:56< boucman> Crab_: rereadh your changes... not much to comment
20090527 19:27:18< Crab_> boucman: ok
20090527 19:28:21< mordante> oh how not surprising r35883 broke something..
20090527 19:28:32< boucman> wesbot: log r35883
20090527 19:28:35< wesbot> suokko * r35883 : Make new turn, turn X, side turn and turn refresh synchronous (bug #10603)
20090527 19:28:38< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=35883
20090527 19:29:02< Ivanovic> hi
20090527 19:29:09< mordante> boucman, FYI Sirp revoked his commit access
20090527 19:29:13< mordante> hi Ivanovic 
20090527 19:29:22< boucman> mordante: ok
20090527 19:39:23-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090527 19:43:02-!- ancestral [n=ancestra@97-116-123-40.mpls.qwest.net] has joined #wesnoth-dev
20090527 19:52:54-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090527 19:53:00-!- Crab_ [n=Crab_@wesnoth/developer/crab] has left #wesnoth-dev []
20090527 19:54:32-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090527 19:57:42-!- YogiHH [n=chatzill@d098122.adsl.hansenet.de] has joined #wesnoth-dev
20090527 19:57:55< YogiHH> hello everyone
20090527 19:58:01< euschn> hi YogiHH 
20090527 19:59:27-!- AnMaster [n=AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20090527 19:59:59< mordante> hi YogiHH 
20090527 20:00:53< loonycyborg> Soliton: I've managed to wmllint scenario attached to bug #10603 but it just goes OOS no matter whether suokko's patch is applied or not.
20090527 20:01:04< loonycyborg> Probably it's just too old..
20090527 20:02:01< Soliton> well, it's pretty simple though isn't it? can you post the linted version?
20090527 20:04:12< Soliton> i'm wondering what version suokko tested on... i mean he must have tested on some thing...
20090527 20:05:07< loonycyborg> Yes. It's a shame that he isn't here on irc anymore..
20090527 20:08:30< loonycyborg> Soliton: I scp'd it to the server to my_home_dir/Test :P
20090527 20:10:39< Soliton> hmm, binary_path wasn't updated but i guess it's not used. wmllint should try to take care of that though...
20090527 20:11:07< loonycyborg> I've also had to comment id tag and add type tags.
20090527 20:11:50< loonycyborg> Perhaps I should have done something more appropriate about missing type tag..
20090527 20:13:13< Soliton> what was wrong with the scenario id?
20090527 20:13:34< loonycyborg> It was causing that spellcheck exception.
20090527 20:14:23< loonycyborg> I didn't find any flags to disable spellcheck in wmllint --help, so..
20090527 20:14:26< Soliton> well, always good to have correctly spelled ids i guess...
20090527 20:16:33< Soliton> there probably should be options in wmllint for all those directives mainline WML is littered with.
20090527 20:17:35< Soliton> so, what OOS did you get?
20090527 20:18:04< Soliton> terrain looked different on different clients?
20090527 20:18:57< loonycyborg> I've got messages about client sending invalid commands.
20090527 20:19:49< Soliton> oh..
20090527 20:21:41< Soliton> hmm, i would have thought the current player does the syncronisation in which case the server shouldn't block it.
20090527 20:23:44< loonycyborg> But indeed clients see different terrain after the OOS.
20090527 20:24:16< Soliton> loonycyborg: can you add '|| command.child("init_side")' to game::is_legal_command() and try again with a recompiled server?
20090527 20:29:04< loonycyborg> I'm still getting OOS, but on turn 2, not turn 1.
20090527 20:29:25< loonycyborg> Still with illegal commands.
20090527 20:30:01< Soliton> what is the illegal command?
20090527 20:31:45< loonycyborg> Really weird thing is happening actually. After the first client ends its turn, it gets to move again with other clients unit :/
20090527 20:31:57< loonycyborg> *client's
20090527 20:33:00< loonycyborg> I'm getting this in output when illegal commands happen:
20090527 20:33:03< loonycyborg> 20090527 22:32:16 error replay: unrecognized action:
20090527 20:33:10< loonycyborg> [choose]
20090527 20:33:16< loonycyborg> value = 0
20090527 20:33:25< loonycyborg> [/choose]
20090527 20:34:52< Soliton> the illegal command is in the server output.
20090527 20:35:55-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090527 20:37:54< loonycyborg> Soliton: http://wesnoth.pastebin.com/d513d3946
20090527 20:38:49< Soliton> hrm
20090527 20:39:32< Soliton> did you have delayd shroud updates on?
20090527 20:40:00< loonycyborg> Nope. I never enable them.
20090527 20:41:24< Soliton> remove the first statement of is_legal_command() then it should work.
20090527 20:43:33-!- YogiHH [n=chatzill@d098122.adsl.hansenet.de] has left #wesnoth-dev []
20090527 20:43:44-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit ["ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]"]
20090527 20:47:15< Soliton> still odd tat you get those errors...
20090527 20:48:29< Soliton> there must have been an unexpected end_turn command somewhere because the current player is allowed to do anything and only the current player should end the turn...
20090527 20:49:20< loonycyborg> I'm still getting [command][end_turn] illegal commands after removing if (!command.one_child()) return false; from is_legal_command.
20090527 20:49:42< Soliton> right...
20090527 20:50:07< Soliton> when does it start?
20090527 20:50:37< loonycyborg> Right after dialog that asks about terrain type.
20090527 20:51:46< Soliton> that means turn 1 side turn 2?
20090527 20:52:25< loonycyborg> It appears after I click end turn twice.
20090527 20:53:07< Soliton> ok, this is all too confusing i guess i have to test myself.. :-P
20090527 20:53:08< loonycyborg> Though both turns are controlled by the client that originally started it.
20090527 20:53:23< Soliton> right, that's not good.
20090527 20:53:43< Soliton> the client is confused then.
20090527 20:54:15< Soliton> you can not end the turn with the second client when it should be his turn
20090527 20:54:17< Soliton> ?
20090527 20:54:43< loonycyborg> end turn is always greyed out for second client.
20090527 20:56:34< Soliton> you could remove the type key from [side] and try again.
20090527 20:57:36< loonycyborg> It doesn't work without type key. That's why I added it.
20090527 20:57:49< loonycyborg> What should I place instead of it?
20090527 20:58:54< Soliton> oh, i looked at a no_leader side.
20090527 20:59:29-!- elynia [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 21:00:31< Soliton> hmm, a new land doesn't have a type key either though.
20090527 21:01:20-!- elynia [n=elynia@wesnoth/bot/shikadibot] has quit [Client Quit]
20090527 21:01:51-!- elynia [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 21:03:47< Soliton> hmm, yeah weird that the client thinks both sides are his.
20090527 21:05:34-!- elynia [n=elynia@wesnoth/bot/shikadibot] has quit [Nick collision from services.]
20090527 21:05:46-!- elynia [n=elynia@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 21:06:29-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has quit ["Lost terminal"]
20090527 21:09:10< CIA-77> mordante * r35912 /trunk/CMakeLists.txt: Add the debug layout option to cmake.
20090527 21:09:14< CIA-77> mordante * r35913 /trunk/src/gui/widgets/panel.cpp: Fix some whitespace issues.
20090527 21:10:27-!- happygrue [n=George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20090527 21:11:54-!- Shadow_Master [n=ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20090527 21:17:31< Soliton> zookeeper: around?
20090527 21:18:50< Soliton> zookeeper: anything looking fishy in this WML? http://nopaste.com/p/au0YT54Ly
20090527 21:19:15< zookeeper> why do people always catch me precisely when i'm about to leave in <10 seconds?
20090527 21:19:42< Soliton> can't catch you when you're gone now can we? :-)
20090527 21:19:51< zookeeper> true enough
20090527 21:19:58< zookeeper> looks fine on a very quick glance
20090527 21:20:06< Soliton> ok, thanks.
20090527 21:20:29< zookeeper> err, except that
20090527 21:20:42< zookeeper> [option]s in any turn events cause OOS
20090527 21:20:50< zookeeper> unless that was fixed, which i doubt
20090527 21:21:06< zookeeper> (away ->)
20090527 21:21:08< Soliton> right, that's what we're trying to fix or rather check if it's fixed.
20090527 21:21:47< Soliton> but the strange part is that if you give both sides to different clients the first one thinkg both sides are his..
20090527 21:25:09< mordante> I'm off now bye
20090527 21:25:32-!- mordante [n=mordante@wesnoth/developer/mordante] has quit ["Leaving"]
20090527 21:29:21-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090527 21:30:40-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 113 (No route to host)]
20090527 21:32:45-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090527 21:33:30< Soliton> loonycyborg: when i try with one client and observe with the other it works at least.
20090527 21:35:08-!- happygrue__ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090527 21:35:24-!- maxy [n=maxy@80-219-0-199.dclient.hispeed.ch] has quit []
20090527 21:35:55-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090527 21:38:59< Soliton> and if i start the game with one client and try to observe after it was started i get a "scenario found: no" error...
20090527 21:39:46-!- happygrue___ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has joined #wesnoth-dev
20090527 21:40:02< Soliton> this is all quite weird. i wonder if it's something special about this scenario or trunk is broken.
20090527 21:42:22-!- happygrue [n=George@wesnoth/developer/wintermute] has quit [Read error: 113 (No route to host)]
20090527 21:42:51-!- cjhopman [n=chris@wesnoth/developer/cjhopman] has quit [Remote closed the connection]
20090527 21:48:24-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 21:51:15< Soliton> ok, after making sure everything is uptodate it seems to work here.
20090527 21:51:32-!- happygrue_ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 113 (No route to host)]
20090527 21:51:56< Soliton> loonycyborg: try with this: http://nopaste.com/p/a7S1rIOQob
20090527 21:52:55< Soliton> i removed some stuff from the side definitions, maybe some of that helped.
20090527 21:53:59-!- happygrue__ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 113 (No route to host)]
20090527 21:55:39< loonycyborg> Soliton: I'm not getting any OOSes with that scenario.
20090527 21:56:12< Soliton> yeah, all the errors where because wesnoth was confused about side ownership.
20090527 21:56:40< Soliton> though, i also found an actual error in the server code. :-)
20090527 22:01:11< Soliton> loonycyborg: if you could isolate what caused that confusion that might be interesting.
20090527 22:02:00< CIA-77> soliton * r35914 /branches/1.6/src/server/game.cpp: removed obsolete check
20090527 22:02:39< CIA-77> soliton * r35915 /trunk/src/server/game.cpp: removed obsolete check
20090527 22:03:36-!- YogiHH [n=chatzill@d028045.adsl.hansenet.de] has joined #wesnoth-dev
20090527 22:07:06< loonycyborg> Hm.. Weird. In your new version of scenario type= is still set to Footpad but it uses player-selected leaders..
20090527 22:08:52< Soliton> i switched the order of side= and type= maybe that has actually an effect?
20090527 22:11:51< loonycyborg> Nope. If I switch them back it still uses player-selected units.
20090527 22:12:25< Shadow_Master> I'm thinking that may be we should split the Mainline Campaign Feedback forum into subforums
20090527 22:12:44-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090527 22:13:28< Shadow_Master> opinions? Turuk, Ivanovic ?
20090527 22:13:44< Ivanovic> Shadow_Master: you mean one per campaign?
20090527 22:13:47< Ivanovic> might make sense
20090527 22:13:47< Shadow_Master> yes
20090527 22:13:54< Shadow_Master> (of course, what else ?)
20090527 22:14:06< Turuk> hmm?
20090527 22:14:09< Turuk> why?
20090527 22:14:25< Shadow_Master> I can only see a handful of topics below the enormous amount of stickies
20090527 22:14:39< Ivanovic> the stickies should be the redirects to the posts
20090527 22:15:05< Ivanovic> the order of the posts *should* not matter, all needed should be the stikies providing a list of threads to post in for the campaign
20090527 22:15:19< Turuk> As long as all the stickies show on the first page, I don't see the issue?
20090527 22:17:04< Turuk> Shadow_Master: What if the campaigns were put into one sticky, that had links to Campaign threads, each of which contained links to the scenarios
20090527 22:18:29< Shadow_Master> um, that may work as well
20090527 22:18:43< Turuk> And just unsticky all but one thread
20090527 22:18:49< Turuk> instead of making all the subforums
20090527 22:18:54< Shadow_Master> (am I the only one who thinks phpbb isn't the best for this stuff anyway?)
20090527 22:27:18-!- happygrue___ [n=George@c-67-176-145-41.hsd1.in.comcast.net] has quit [Read error: 110 (Connection timed out)]
20090527 22:47:42-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit ["The man who smiles when things go wrong has thought of someone to blame it on."]
20090527 22:50:06-!- ancestral [n=ancestra@97-116-123-40.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090527 22:52:45< CIA-77> soliton * r35916 /trunk/utils/mp-server/ (run_server update_server): various tweaks/fixes after testing
20090527 22:58:47-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090527 23:07:49-!- elynia [n=elynia@wesnoth/bot/shikadibot] has quit ["."]
20090527 23:08:47-!- elynia [n=shyde@wesnoth/bot/shikadibot] has joined #wesnoth-dev
20090527 23:12:48-!- elynia [n=shyde@wesnoth/bot/shikadibot] has quit [Client Quit]
20090527 23:13:01< Soliton> can anyone understand and reproduce this report? https://gna.org/bugs/?13577
20090527 23:15:10< Shadow_Master> "open then cancel" . Well, doing it exactly as it says there doesn't reproduce the problem.
20090527 23:15:49< Soliton> not that he says what to open.. but i guess he means the load game dialog?
20090527 23:16:02< Noyga> n8
20090527 23:16:03-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit ["Quitte"]
20090527 23:16:16< Soliton> that indeed works fine here, too.
20090527 23:16:47< Shadow_Master> I can't think of/find anything else that could possibly be involed with reloading an autosave
20090527 23:16:58< Shadow_Master> reloading the autosave the proper way doesn't crash either
20090527 23:17:54< Soliton> and there is no other dialog involved unless the save is from a different version. so i guess it is a "works for me".
20090527 23:18:04-!- elynia [n=shyde@146.155.83.105] has joined #wesnoth-dev
20090527 23:18:45< YogiHH> Soliton: did you try loading from multiplayer, too?
20090527 23:19:32< Soliton> only
20090527 23:19:48< Shadow_Master> I was trying from single player, HttT 1
20090527 23:19:51< Soliton> oh, you mean from create game?
20090527 23:19:58< YogiHH> yes
20090527 23:19:58< Shadow_Master> turn 1 or turn 2. Neither crashes.
20090527 23:20:43< Soliton> YogiHH: doesn't really fit the description since he want's to get back to the current game.
20090527 23:21:09< Soliton> works fine also.
20090527 23:22:13< Shadow_Master> perhaps someone could try on the specified platform, which is "X 10.5.7 (9J61)". Can't mean anything else than MacOS X hpoefully
20090527 23:24:11< Soliton> hmm, i managed again to produce a save with no gamestate or something though.
20090527 23:24:56< YogiHH> Soliton: what kind of save, start-of-scenario?
20090527 23:27:10< Soliton> YogiHH: mp
20090527 23:27:31< Soliton> it actually does load though.
20090527 23:27:56< Soliton> but the name is just autosave.gz
20090527 23:28:18< YogiHH> hmm, weird
20090527 23:29:20< Soliton> http://nopaste.com/p/a8LMueIwqb
20090527 23:30:25< Soliton> i wonder how you can produce empty [command]s.
20090527 23:32:00-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has left #wesnoth-dev []
20090527 23:33:15-!- giusef [n=giusef@unaffiliated/giusef] has quit ["exit (-1);"]
20090527 23:35:38< YogiHH> Soliton: for some reason, there is a lot of root-level information missing (label, completion, campaign_type). As the label is used for building the autosave name, it explains what happens (though not why there is no label).
20090527 23:36:28< Soliton> AI0867: could you check whether https://gna.org/bugs/index.php?13523 was fixed?
20090527 23:38:17< Soliton> YogiHH: seems i can reproduce by loading from a turn 1 autosave.
20090527 23:39:07< YogiHH> Soliton: so the first save has a label and the second does not?
20090527 23:39:19< Soliton> yep, now even that is the case.
20090527 23:39:52< YogiHH> can you paste the first save, please?
20090527 23:40:18< Soliton> http://nopaste.com/p/aJhmJNzC4
20090527 23:41:08< YogiHH> yes, that's what it is supposed to look like
20090527 23:41:40< Soliton> yeah and when i load that again it works...
20090527 23:41:48-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090527 23:42:23< YogiHH> i wonder if it has to do with this line in savegame.cpp.write_game_snapshot: 	snapshot().merge_attributes(level_cfg_);
20090527 23:42:38< YogiHH> but i doubt it
20090527 23:42:59-!- elynia is now known as elynia_
20090527 23:43:20-!- elynia_ [n=shyde@146.155.83.105] has left #wesnoth-dev []
20090527 23:43:55< Soliton> again reproduced on trunk.
20090527 23:44:11< YogiHH> anyway, i need to look for this another time
20090527 23:44:15< Soliton> first autosave is fine next is messed. no reloading needed.
20090527 23:44:16< YogiHH> good night everyone
20090527 23:44:20< Soliton> night.
20090527 23:44:48-!- YogiHH [n=chatzill@d028045.adsl.hansenet.de] has left #wesnoth-dev []
20090527 23:47:09< Soliton> oops, that was still 1.7.0...
20090527 23:47:11-!- elynia [n=shyde@146.155.83.105] has joined #wesnoth-dev
20090527 23:47:58< Shadow_Master> um, that may be it. I tried with trunk r35884 and nothing happened. Then again it was a single player game
20090527 23:48:06< Soliton> ah well, still reproduced with trunk.
20090527 23:48:47< Soliton> r35909
20090527 23:53:58< Shadow_Master> oh great, git decided to recompress objects just when I was about to shutdown and leave the lab.
20090527 23:54:46< Soliton> so cancel it.
20090527 23:54:54< Shadow_Master> it won't corrupt anything?
20090527 23:55:18< Soliton> no...
20090527 23:55:21-!- elynia [n=shyde@146.155.83.105] has quit ["Lost terminal"]
20090527 23:57:08-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090527 23:57:54-!- Elvish_Pillage2 [n=eli@68.116.173.131] has joined #wesnoth-dev
20090527 23:58:11-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit [Client Quit]
--- Log closed Thu May 28 00:00:11 2009