--- Log opened Wed Apr 28 00:00:42 2010
20100428 00:03:57-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100428 00:05:23< crimson_penguin> this is a little bit hard to look into; lots of variables coming from all over the place
20100428 00:05:58< Ivanovic> i thought you would like to have some real challange to face
20100428 00:05:59< Ivanovic> ;)
20100428 00:06:29< crimson_penguin> oh, I think I see the problem
20100428 00:06:39< crimson_penguin> I made a deduction that narrowed it down a bit ;)
20100428 00:08:17< crimson_penguin> ...it's snowing here
20100428 00:13:34< Ivanovic> poor crimson_penguin
20100428 00:13:39< Ivanovic> anyway, really off to bed now
20100428 00:22:35-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Quit: Hanging with Aaron.]
20100428 00:23:12-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100428 00:32:44-!- YogiHH [YogiHH@wesnoth/developer/yogihh] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100428 00:37:16-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100428 00:37:26-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100428 00:38:24-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Client Quit]
20100428 00:42:37-!- Skystriker [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has joined #wesnoth-dev
20100428 00:49:04-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 258 seconds]
20100428 00:52:12-!- Hethrir [~chatzilla@adsl-75-49-234-30.dsl.stlsmo.sbcglobal.net] has joined #wesnoth-dev
20100428 00:52:50-!- k23z__ [~k23z__@unaffiliated/k23z--/x-2536701] has joined #wesnoth-dev
20100428 00:52:54< Hethrir> Hello
20100428 00:53:03< Hethrir> Dang Crab_ is gone :(
20100428 00:53:38-!- Bob_The_Mighty [~chatzilla@cpc4-brig15-0-0-cust904.3-3.cable.virginmedia.com] has joined #wesnoth-dev
20100428 00:59:13-!- mrEPIC [~mrEPIC@adsl-76-199-72-45.dsl.hstntx.sbcglobal.net] has quit [Quit: Leaving]
20100428 01:02:54-!- gabba [~gabriel@70.35.167.54] has joined #wesnoth-dev
20100428 01:05:37< Hethrir> Hey gabba
20100428 01:05:47< gabba> hey Hethrir
20100428 01:08:18-!- ilor_ [~ilor@wesnoth/developer/ilor] has joined #wesnoth-dev
20100428 01:10:55-!- ilor__ [~ilor@auk24.neoplus.adsl.tpnet.pl] has quit [Ping timeout: 258 seconds]
20100428 01:22:05-!- elias [~elias@allegro/developer/allefant] has quit [Quit: Leaving]
20100428 01:31:14-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has joined #wesnoth-dev
20100428 01:34:17-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100428 01:40:33-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has joined #wesnoth-dev
20100428 01:40:33-!- stikonas [~and@bcm-131-111-247-104.girton.cam.ac.uk] has quit [Changing host]
20100428 01:40:33-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100428 01:42:03< CIA-10> crimson_penguin * r42289 /website/gettext.wesnoth.org/index.php: Hopefully fixed sorting by team name.
20100428 01:42:30 * crimson_penguin doesn't know how to make it update
20100428 01:42:39< crimson_penguin> Ivanovic: see if that fixes it
20100428 01:43:48-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has joined #wesnoth-dev
20100428 01:50:38-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100428 01:50:59-!- wesbot changed the topic of #wesnoth-dev to: 98 bugs, 273 feature requests, 18 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100428 01:55:14-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 01:56:40< Hethrir> Hey Elendil
20100428 01:57:23< King_Elendil> hey :)
20100428 01:58:15-!- Hethrir [~chatzilla@adsl-75-49-234-30.dsl.stlsmo.sbcglobal.net] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100428 02:00:55-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100428 02:01:39-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 02:04:11-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100428 02:16:10-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100428 02:16:33-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 02:19:59-!- fakedrake [~fakedrake@ppp-94-64-201-117.home.otenet.gr] has joined #wesnoth-dev
20100428 02:27:52< fakedrake> hello
20100428 02:29:33< King_Elendil> hi
20100428 02:30:45-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100428 02:31:26-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 02:44:31-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100428 02:45:36-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 258 seconds]
20100428 02:46:17-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100428 02:46:21-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 02:53:10-!- k23z__ [~k23z__@unaffiliated/k23z--/x-2536701] has quit [Quit: Leaving]
20100428 03:00:17-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100428 03:06:46-!- Upthorn [ogmar@adsl-75-26-172-81.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 245 seconds]
20100428 03:11:55-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100428 03:12:34-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100428 03:15:24-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Read error: Connection reset by peer]
20100428 03:16:01-!- King_Elendil [~King_Elen@75.143.233.138] has joined #wesnoth-dev
20100428 03:31:53-!- fakedrake [~fakedrake@ppp-94-64-201-117.home.otenet.gr] has quit [Ping timeout: 265 seconds]
20100428 03:34:52-!- King_Elendil is now known as King_Elendil_
20100428 03:37:14-!- King_Elendil_ is now known as King_Elendil
20100428 03:37:26-!- gabba [~gabriel@70.35.167.54] has left #wesnoth-dev []
20100428 03:39:40-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100428 03:42:01-!- King_Elendil [~King_Elen@75.143.233.138] has quit [Quit: I hope y'all have a nice day ;)]
20100428 04:03:00-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100428 04:33:13-!- Ivanovic_ [~ivanovic@dtmd-4db22c4b.pool.mediaWays.net] has joined #wesnoth-dev
20100428 04:36:21-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 245 seconds]
20100428 04:36:48-!- Blueblaze [~nick@adsl-76-202-23-150.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100428 04:37:11-!- Ivanovic_ is now known as Ivanovic
20100428 04:48:27-!- Bob_The_Mighty [~chatzilla@cpc4-brig15-0-0-cust904.3-3.cable.virginmedia.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]]
20100428 04:50:46-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has left #wesnoth-dev []
20100428 05:08:42-!- Upthorn [~ogmar@adsl-75-26-172-81.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100428 05:20:46-!- Blarumyrram [~Blarumyrr@81-20-159-197.levira.ee] has quit [Quit: Lahkun]
20100428 05:30:51-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100428 05:39:51-!- dtiger [~dtiger@dynamic-vpdn-91-149-132-58.telecom.by] has joined #wesnoth-dev
20100428 06:00:49-!- Skystriker [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has quit [Ping timeout: 246 seconds]
20100428 06:12:05-!- dtiger [~dtiger@dynamic-vpdn-91-149-132-58.telecom.by] has quit [Remote host closed the connection]
20100428 06:18:32-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100428 06:28:01-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100428 06:32:38-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100428 06:39:09-!- noy_ [~Noy@d99-199-3-111.bchsia.telus.net] has joined #wesnoth-dev
20100428 06:39:09-!- noy_ [~Noy@d99-199-3-111.bchsia.telus.net] has quit [Changing host]
20100428 06:39:09-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100428 06:39:42-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: Connection reset by peer]
20100428 06:39:42-!- noy_ is now known as noy
20100428 07:11:27-!- mordocai [~mordocai@66.119.9.243] has joined #wesnoth-dev
20100428 07:30:23-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100428 07:53:20-!- mordocai [~mordocai@66.119.9.243] has quit [Quit: Leaving]
20100428 07:57:37-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100428 08:33:47-!- halifix [~halifix@adsl-69-109-125-143.dsl.pltn13.pacbell.net] has quit [Quit: (five]
20100428 08:46:29-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100428 09:16:43-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100428 09:21:51-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100428 09:22:38-!- euschn [~eugen@wesnoth/developer/euschn] has joined #wesnoth-dev
20100428 09:25:07-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100428 09:31:39-!- Netsplit *.net <-> *.split quits: esr, Bocom
20100428 09:37:36-!- Netsplit over, joins: esr, Bocom
20100428 09:38:26-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20100428 09:38:26-!- esr [~chatzilla@wesnoth/developer/esr] has joined #wesnoth-dev
20100428 10:03:35-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100428 10:03:55-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100428 10:14:29-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100428 10:25:43-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100428 10:30:21-!- Darkas [~quassel@ppp-93-104-134-176.dynamic.mnet-online.de] has joined #wesnoth-dev
20100428 10:36:03-!- Darkas [~quassel@ppp-93-104-134-176.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100428 10:38:22-!- ilor_ [~ilor@wesnoth/developer/ilor] has quit [Read error: Connection reset by peer]
20100428 10:41:57-!- Ivanovic [~ivanovic@dtmd-4db22c4b.pool.mediaWays.net] has quit [Changing host]
20100428 10:41:57-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100428 10:42:30< Ivanovic> hi
20100428 10:45:10< Crab_> hi, Ivanovic
20100428 10:46:31-!- fendrin [~fabi@77-20-109-65-dynip.superkabel.de] has joined #wesnoth-dev
20100428 10:46:31-!- fendrin [~fabi@77-20-109-65-dynip.superkabel.de] has quit [Changing host]
20100428 10:46:31-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100428 10:51:53-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100428 11:00:02-!- euschn [~eugen@wesnoth/developer/euschn] has quit [Ping timeout: 240 seconds]
20100428 11:19:15-!- Darkas [~quassel@ppp-93-104-134-176.dynamic.mnet-online.de] has joined #wesnoth-dev
20100428 11:37:31-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100428 11:38:33< timotei> hey Ivanovic 
20100428 11:38:42< Ivanovic> timotei: you read the logs?
20100428 11:38:53< timotei> Right now, I've seen a message from you
20100428 11:38:57< Ivanovic> :)
20100428 11:40:38< timotei> Ivanovic, so, I should update the abstract/title right?
20100428 11:40:48< Ivanovic> timotei: once it is possible: yes, please
20100428 11:40:48< timotei> so it will reflect on what I'm working
20100428 11:40:53< Ivanovic> currently it does suck in those regards
20100428 11:40:57< timotei> I can edit it now
20100428 11:41:02< Ivanovic> cool
20100428 11:41:23< Ivanovic> if you just got to http://socghop.appspot.com/gsoc/org/home/google/gsoc2010/wesnoth and look at the list of proposals you see what i mean
20100428 11:41:30< timotei> ok
20100428 11:41:38< timotei> should I "update" my location too?
20100428 11:41:40< Ivanovic> in general visitors of this page are meant to get a good idea about the proposals
20100428 11:41:46< timotei> so it can be seen the map?
20100428 11:41:47 * Ivanovic does not care much about the location
20100428 11:41:59< Ivanovic> you are free to do so if you want to, but it is not mandatory
20100428 11:42:24< Ivanovic> having a good title and abstract though (maybe include a link to you wiki page for further details) is something really usefull
20100428 11:42:33-!- Darkas [~quassel@ppp-93-104-134-176.dynamic.mnet-online.de] has quit [Remote host closed the connection]
20100428 11:43:00< timotei> ok
20100428 11:46:53< timotei> ok Ivanovic, done that:)
20100428 11:46:59< timotei> tell me if it it ok
20100428 11:47:02< timotei> if it is
20100428 11:48:06< Ivanovic> timotei: sounds good to me
20100428 11:48:10< timotei> ok
20100428 11:48:37< Ivanovic> billynux, gabba, Upthorn: if you want to update your entries title/abstract: should be possible now, at least timotei was able to do so
20100428 11:51:40-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100428 11:51:59-!- Blueblaze [~nick@adsl-76-202-23-150.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
20100428 11:57:16-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has joined #wesnoth-dev
20100428 11:57:16-!- loonybot [~loonybot@ppp79-139-136-187.pppoe.spdop.ru] has quit [Changing host]
20100428 11:57:16-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100428 11:57:30< timotei> ok, I'm leaving for now. See you later
20100428 11:57:36-!- timotei [~timotei@193.34.191.4] has quit [Quit: Leaving]
20100428 11:58:09-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100428 11:58:59-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100428 12:05:39-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Quit: Leaving]
20100428 12:18:56-!- shikadib1t [~shikadi@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev
20100428 12:19:03< CIA-10> ivanovic * r42290 /trunk/po/ (4 files in 4 dirs): updated German translation
20100428 12:19:05< CIA-10> ivanovic * r42291 /branches/1.8/po/ (4 files in 4 dirs): updated German translation
20100428 12:24:16-!- Netsplit *.net <-> *.split quits: shikadibot
20100428 12:53:44-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has joined #wesnoth-dev
20100428 13:31:13-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100428 13:32:09-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100428 13:50:59-!- wesbot changed the topic of #wesnoth-dev to: 99 bugs, 273 feature requests, 18 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100428 13:53:53-!- DesertPanther [~Khalid@41.234.148.141] has joined #wesnoth-dev
20100428 13:53:58-!- DesertPanther [~Khalid@41.234.148.141] has quit [Changing host]
20100428 13:53:58-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100428 13:55:49-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Client Quit]
20100428 13:56:48-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Read error: Connection reset by peer]
20100428 13:57:20< CIA-10> zookeeper * r42292 /branches/1.8/ (3 files in 2 dirs): Made the final objective in 'Vengeance' be revealed immediately at the beginning of the scenario.
20100428 14:07:41-!- euschn [~eugen@wesnoth/developer/euschn] has joined #wesnoth-dev
20100428 14:08:34< shadowmaster> Greywhind: what's your forum account?
20100428 14:08:49-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100428 14:09:33< Ivanovic> timotei: what is the status of these patches submitted by you: https://gna.org/patch/?1644 and https://gna.org/patch/?1625
20100428 14:09:45< Ivanovic> are they commited, do you want some more review, ...?
20100428 14:09:46< CIA-10> zookeeper * r42293 /trunk/ (3 files in 2 dirs): Ported r42292 to trunk.
20100428 14:11:57-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100428 14:13:08-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100428 14:16:40-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100428 14:16:42-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100428 14:21:53< shadowmaster> I
20100428 14:25:04-!- Skystriker [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has joined #wesnoth-dev
20100428 14:25:04-!- Skystriker is now known as Skywork
20100428 14:25:19< shadowmaster> I like the new animated water. Too bad it drops the framerate by 20 :/
20100428 14:28:38-!- Appleman1234 [~Appleman1@CPE-124-191-176-143.oxqn1.cha.bigpond.net.au] has quit [Ping timeout: 258 seconds]
20100428 14:41:18-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100428 14:56:38-!- Skywork [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has quit [Ping timeout: 252 seconds]
20100428 14:58:58-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100428 15:16:39-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100428 15:17:33-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has left #wesnoth-dev []
20100428 15:20:46-!- Aizu` [~drusepth@dhcp-077-249-151-209.chello.nl] has quit [Ping timeout: 258 seconds]
20100428 15:24:01-!- shikadib1t is now known as shikadibot
20100428 15:30:52< Ivanovic> Crab_: since those patches are assigned to you:
20100428 15:30:53< Ivanovic> [14:09:33] <Ivanovic> timotei: what is the status of these patches submitted by you: https://gna.org/patch/?1644 and https://gna.org/patch/?1625
20100428 15:31:38< Crab_> looking...
20100428 15:32:56< Crab_> https://gna.org/patch/?1625 was ok, so it can be committed. https://gna.org/patch/?1644 is still incomplete\ in progress, afair.
20100428 16:24:11-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100428 16:28:27-!- shadowmaster_ [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100428 16:28:32-!- shadowmaster_ is now known as shadowm_bluecore
20100428 16:30:45-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100428 16:34:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 264 seconds]
20100428 16:34:14< shadowmaster> Ivanovic: have the mesa requirements changed lately?
20100428 16:34:49< Ivanovic> huh?
20100428 16:35:03< Ivanovic> i don't think that stuff changed there lately
20100428 16:35:06< shadowmaster> lately as in: since February
20100428 16:35:22< Ivanovic> though what is relevant for mesa is the xf86-video driver as well as libdrm to be recent enough
20100428 16:35:36< shadowmaster> yeah, but that's not the problem. It's a compile-time issue
20100428 16:35:38< Ivanovic> (that being the reason why i run the git version of those three)
20100428 16:35:41< shadowmaster> and it has something to do with llvm
20100428 16:36:00< Ivanovic> llvm is some kind of software render modus
20100428 16:36:08< Ivanovic> you can deactivate it for your needs
20100428 16:36:25< Ivanovic> you can even deactivate all of gallium since your r600 based card has no gallium driver so far
20100428 16:36:36< shadowmaster> well, I did pass --disable-gallium to the configure script
20100428 16:37:04< shadowmaster> CFLAGS='-O3 -mtune=native -march=native' CXXFLAGS='-O3 -mtune=native -march=native' ./configure --prefix=/usr --enable-glx-tls --with-dri-drivers=radeon,r600,swrast --disable-gallium
20100428 16:38:10< shadowmaster> and yet: http://pastebin.com/tYxnWXTY
20100428 16:38:42< Ivanovic> ./configure --prefix=/usr --build=x86_64-pc-linux-gnu --host=x86_64-pc-linux-gnu --mandir=/usr/share/man --infodir=/usr/share/info --datadir=/usr/share --sysconfdir=/etc --localstatedir=/var/lib --libdir=/usr/lib64 --disable-option-checking --with-driver=dri --disable-glut --without-demos --disable-debug --disable-glw --disable-motif --enable-glx-tls --enable-xcb --enable-asm --with-dri-drivers=,swrast,radeon,r200,r300,r600 --disable-
20100428 16:38:43< Ivanovic> gallium
20100428 16:38:46< shadowmaster> um, that "-DHAVE_LLVM=0x0@LLVM_VERSION@" parameter looks like a bug inthe recipe
20100428 16:38:50< Ivanovic> that is my configure line as gentoo does use it
20100428 16:39:22< shadowmaster> although now that I think of it, I forgot to run autogen.sh first
20100428 16:39:33< Ivanovic> and yeah, mesa head does not compile for me at the moment
20100428 16:39:47< shadowmaster> yeah, that was it.
20100428 16:42:09< Ivanovic> with other words: you should stop by in one of the open source driver chans (no idea if there is #mesa or if you should just use #radeon)
20100428 16:43:06< shadowmaster> no #mesa
20100428 16:57:05-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100428 16:57:12< timotei> Ivanovic, https://gna.org/patch/?1625 was commited
20100428 16:57:26< timotei> but someone said it shouldn't be closed, until 1.9.0 is shipped
20100428 16:57:35< timotei> I don't really remember who
20100428 16:57:54< timotei> Ivanovic, and yes, the other patch is work in progress
20100428 16:58:51< shadowmaster> it shouldn't be closed, but it should be marked as "done" if applied then
20100428 16:59:24< shadowmaster> (closing a tracker entry isn't the same as marking it as Fixed or Done)
20100428 16:59:24-!- euschn [~eugen@wesnoth/developer/euschn] has quit [Quit: Leaving.]
20100428 16:59:36< timotei> ok, shadowmaster , thanks
20100428 17:00:12< timotei> I am allowed to mark it as Done? or only Crab_ can do this? (in the way that: Crab_ is assigned to)
20100428 17:03:43< shadowmaster> if you are sure it was applied correctly, sure
20100428 17:04:02< timotei> ok
20100428 17:04:18< timotei> I'll let Crab_ do that then, just to be sure there isn't any problem
20100428 17:04:46< Crab_> timotei: mark it as done :)
20100428 17:04:50< timotei> ok
20100428 17:04:53< Crab_> thanks
20100428 17:05:03< timotei> np
20100428 17:05:38-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100428 17:11:40-!- deekay [~dk@wesnoth/developer/dragonking] has joined #wesnoth-dev
20100428 17:24:20-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100428 17:30:26-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100428 17:37:17< shadowmaster> Ivanovic: it should compile now (git master, that is)
20100428 17:39:51-!- shadowm_bluecore [~ignacio@wesnoth/developer/shadowmaster] has quit [Quit: restarting X]
20100428 17:44:39-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100428 17:47:46< Ivanovic> timotei: but it should be marked as "done"
20100428 17:48:00-!- shadowm_bluecore [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100428 17:48:29-!- elias [~elias@allegro/developer/allefant] has joined #wesnoth-dev
20100428 17:49:38-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Read error: Connection reset by peer]
20100428 17:50:32< Ivanovic> shadowmaster: yeah, it does build now
20100428 17:52:21-!- timotei [~timotei@193.34.191.4] has quit [Quit: brb]
20100428 17:55:16-!- Sirp__ [~97c1dc1c@wesnoth/developer/dave] has joined #wesnoth-dev
20100428 17:58:14-!- Johannes13__ [~Johannes@pD95005A2.dip0.t-ipconnect.de] has joined #wesnoth-dev
20100428 18:01:07-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 246 seconds]
20100428 18:04:06< shadowmaster> I had not noticed that Glamdrol uses Junicode
20100428 18:05:00< shadowmaster> for h1, h2 and h3 elements. Of course, since I didn't have Junicode installed before...
20100428 18:06:17-!- Lastmerlin [~Lastmerli@kalypso.csn.tu-chemnitz.de] has joined #wesnoth-dev
20100428 18:06:49< Greywhind> shadowmaster: my forum name is Greywhind
20100428 18:07:44< shadowmaster> and you are a developer now
20100428 18:08:41< Greywhind> thanks
20100428 18:20:53-!- Blarumyrran [~Blarumyrr@81-20-159-197.levira.ee] has joined #wesnoth-dev
20100428 18:25:52-!- shadowm_bluecore [~ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100428 18:26:04-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has quit [Quit: leaving]
20100428 18:26:19-!- shadowm_bluecore [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100428 18:29:29-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100428 18:33:20-!- timotei [timotei@188.24.6.152] has joined #wesnoth-dev
20100428 18:33:38-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100428 18:37:39-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has joined #wesnoth-dev
20100428 18:41:32-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100428 18:45:42-!- qemqemqem_ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Read error: Connection reset by peer]
20100428 18:46:19-!- Crab_ [~Crab@wesnoth/developer/crab] has joined #wesnoth-dev
20100428 18:46:34-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100428 18:49:40-!- yann__ [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100428 18:50:06-!- yann__ is now known as yann
20100428 18:51:45-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100428 18:52:10-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has joined #wesnoth-dev
20100428 18:52:34< norbert_> hi, is there a RFC about Wesnoth client/server communication?
20100428 18:53:18< norbert_> documentation, that explains how the handshake is done, what packages with what sizes are expected, and so on
20100428 18:53:21< shadowmaster> for the IETF? I don't think so
20100428 18:53:49-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100428 18:53:49< shadowmaster> client-server communication is done via WML and I believe there's some info in the  wiki
20100428 18:54:01< Ivanovic> no rfc
20100428 18:54:03< shadowmaster> nowadays WML streams are compressed as gzip too
20100428 18:54:13< norbert_> I found this http://wiki.wesnoth.org/WesnothdDesign#The_network_layer
20100428 18:54:16< norbert_> but that's kind of limited
20100428 18:54:26< Ivanovic> norbert_: btw, since you were interested in stats stuff: https://mail.gna.org/public/wesnoth-dev/2010-04/msg00043.html
20100428 18:55:39< Ivanovic> (though those were (singleplayer) campaignspecific statistics like number of turns required to finish the scenario, gold at end of scenario, gold at start of scenario, number of leveled up units, ...)
20100428 18:55:56-!- timotei [timotei@188.24.6.152] has quit [Read error: Connection reset by peer]
20100428 18:56:01-!- timotei [timotei@188.24.6.152] has joined #wesnoth-dev
20100428 18:56:40< norbert_> so you're removing stats that I would like to use?
20100428 18:57:34< Crab_> norbert_: they are to be added back in a refactored way.
20100428 18:57:39< norbert_> "currently noone maintaining the campaign statistics anymore"
20100428 18:57:52< norbert_> ok...
20100428 18:58:00< Crab_> norbert_: so, we are to figure out 'what stats to use' and then they'll be readded.
20100428 18:58:18< Ivanovic> they are only added back if we have someone maintaining front and backend of the stats
20100428 18:58:42< norbert_> all I need is a file with stats
20100428 18:58:51< norbert_> it doesn't matter what format, because I can parse pretty much everything
20100428 18:59:10< Ivanovic> norbert_: the stats we have is a rather huge sql database
20100428 18:59:12< norbert_> if we can get it to me daily, I'll handle the frontend
20100428 18:59:22< Ivanovic> and does probably *not* cover the stats you do want to have...
20100428 19:00:05< norbert_> what is "huge"?
20100428 19:00:21< Ivanovic> once upon the time it had 2gb
20100428 19:00:36< norbert_> okay; but that's not a problem
20100428 19:00:36-!- timotei [timotei@188.24.6.152] has quit [Client Quit]
20100428 19:01:32< norbert_> anyways, is there someone here with knowledge about the client/server stuff?
20100428 19:01:43< norbert_> the details, I mean
20100428 19:01:53< norbert_> who could explain to me what is being sent back and forth, and how
20100428 19:02:01< Ivanovic> no, there is *nobody*
20100428 19:02:12< Ivanovic> that is why it is unmaintained!
20100428 19:02:29< Crab_> norbert_: when SP games take place, information is collected
20100428 19:02:46< Ivanovic> though it looks like we once purged the database and are close to zero at the moment regarding the database stuff
20100428 19:02:47< norbert_> I mean in general, when people connect to server.wesnoth.org
20100428 19:02:57< norbert_> the communication from/to port 15000
20100428 19:02:58< Ivanovic> there are no stats for the mp server!
20100428 19:03:12< Ivanovic> that is: no stats beside what you can view at wesnothd.wesnoth.org
20100428 19:04:06< Crab_> Ivanovic: and http://replays.wesnoth.org, as well, isn't it ?
20100428 19:04:36< Ivanovic> Crab_: those are replays
20100428 19:04:41< Ivanovic> not stats that we do gather directly
20100428 19:05:06< Crab_> Ivanovic: well, all the stats can be datamined from the replays, quite easily
20100428 19:05:28< Ivanovic> what about the stuff like "aborted games"?
20100428 19:06:17< Crab_> Ivanovic: they are written there, too ?
20100428 19:06:33< Ivanovic> anyway, those are only the games that are observable
20100428 19:06:39< Ivanovic> many people on the server are not
20100428 19:07:04< Crab_> well, that's ok. after all , it's ok if a not observable game is 'private'
20100428 19:09:05< Crab_> norbert_: so, about SP stuff: when a user plays the game, at any possible action, we can 'log' it to the game stats. for example, his gold, units recruited, units killed, etc. even info about 'where' a unit was killed. all sorts of details.
20100428 19:09:41< Crab_> norbert_: and then, 'somewhen after the SP game is over', that log is sent to wesnoth.org stats collector
20100428 19:10:02< Crab_> norbert_: afaik, it just does some basic sanity checking, and stores things in the DB
20100428 19:12:10-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100428 19:12:44< Crab_> norbert_: take a look at http://wiki.wesnoth.org/SoC_Ideas_Stats_Server
20100428 19:12:47< Crab_> norbert_: it is from 2009
20100428 19:13:42< norbert_> so those tables are already being filled by the server?
20100428 19:13:51< Crab_> norbert_: yes
20100428 19:14:05< Crab_> norbert_ : it was stable for a long time
20100428 19:14:24< Crab_> norbert_: it's just that the web interface rewrite failed.
20100428 19:14:48< Crab_> norbert_: if you know python, you can look at the code which was made by corn in 2009
20100428 19:14:51-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has quit [Ping timeout: 260 seconds]
20100428 19:14:54< norbert_> how much data do you think is added daily?
20100428 19:15:05-!- Blueblaze [~nick@adsl-76-202-23-150.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100428 19:15:16< norbert_> I mean the size, MBs
20100428 19:15:28< Crab_> norbert_: no idea and now I'm away from my ssh key to login to w.o and check.
20100428 19:16:33-!- Johannes13__ [~Johannes@pD95005A2.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds]
20100428 19:16:43< Crab_> norbert_: see http://code.google.com/p/google-summer-of-code-2009-wesnoth/downloads/list
20100428 19:16:50< Crab_> http://google-summer-of-code-2009-wesnoth.googlecode.com/files/Gregory_Shikhman.tar.gz
20100428 19:17:29-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100428 19:17:39< boucman> hey all
20100428 19:17:50< Crab_> norbert: (it's a bunch of diffs,  so it might be easier to just check the final results in wesnoth's svn)
20100428 19:17:53< Crab_> hi, boucman
20100428 19:17:59< norbert_> yes, the diffs aren't very useful
20100428 19:19:03-!- apoi [~andi@85-126-180-242.volume.xdsl-line.inode.at] has joined #wesnoth-dev
20100428 19:19:10< Crab_> http://svn.gna.org/viewcvs/wesnoth/website/stats.wesnoth.org/
20100428 19:19:10-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100428 19:19:24< norbert_> but I would build my own thing anyways
20100428 19:19:33< norbert_> if you can tell me how much MB is added each day
20100428 19:19:47< norbert_> if it's not too much, we can arrange something to send anonymous reports to me
20100428 19:19:58< norbert_> and then I'll just write a frontend that displays interesting data
20100428 19:20:19< Crab_> norbert_: yes, that's a good idea
20100428 19:20:33< norbert_> all open source/free software, of course
20100428 19:20:41< Crab_> norbert_: maybe stop the uploader removal, then ?
20100428 19:20:58< norbert_> yes, that didn't sound good
20100428 19:21:14< norbert_> removing it just when there's someone willing and wanting to process it :)
20100428 19:22:17< Crab_> ok, then I'll send a nack to wesnoth-dev ML in a 1-2 hours (or you can join the list and say for yourself)
20100428 19:22:27< Crab_> wesbot: seen billynux
20100428 19:22:28< wesbot> Crab_: The person with the nick billynux last spoke 20h 28m ago.  20h 26m ago they left with the message: Quit: Page closed
20100428 19:22:33< Crab_> wesbot: seen mordante
20100428 19:22:34< wesbot> Crab_: The person with the nick mordante last spoke 20h 43m ago.  20h 42m ago was here and on the channel #wesnoth-de with the message: Quit: Leaving
20100428 19:22:59< norbert_> if you could do that, thanks
20100428 19:23:26-!- Sirp__ [~97c1dc1c@wesnoth/developer/dave] has quit [Ping timeout: 252 seconds]
20100428 19:23:53< Crab_> billynux, mordante:  it will be a good thing to not remove the stats uploader stuff for now, as norbert_ wants to code a frontend for it, to actually display useful data from it.
20100428 19:24:08< Crab_> norbert_: yes, I'll do it as soon as I return home.
20100428 19:24:34< Crab_> Ivanovic: ^
20100428 19:27:15< norbert_> first I'm going to study static server_type open_connection
20100428 19:27:20< norbert_> and write a RFC
20100428 19:27:35< norbert_> I want a bot on the server that can handle wsrsw vote requests
20100428 19:28:09< norbert_> so people can simply message or say to wsrsw "wsrsw: <map> <rating stuff>" and it processes it
20100428 19:28:34< Crab_> norbert_: it is a irc bot for wesnothd ?
20100428 19:28:43< Crab_> (rc-style)
20100428 19:28:47< norbert_> something like that, I guess
20100428 19:28:56< Aethaeryn> Crab_: Will it be possible for one player (the host for instance) to be on multiple maps at once?
20100428 19:29:17< Aethaeryn> or at least to be able to switch between the maps going on during his turn or something?
20100428 19:30:03< Crab_> norbert_: I will want to write a similar bot-thing for the gamemaster
20100428 19:30:13< Aethaeryn> like if he has many units on multiple maps?
20100428 19:30:45< Crab_> Aethaeryn: I was thinking about it. it is a potential possibility, but it is a separate project.
20100428 19:31:05< Aethaeryn> bah
20100428 19:31:09< Crab_> Aethaeryn: we will complete all the network pre-reqs for it with our new network stack.
20100428 19:31:25< Crab_> Aethaeryn: and, Gamemaster will definitely be able to be on several maps at once
20100428 19:31:39< Crab_> Aethaeryn: but, for normal player, there are ui issues to consider.
20100428 19:31:45< Aethaeryn> Crab_: Can I show you the kind of map I'd be interested in turning into an MP campaign (on 1.8)?
20100428 19:31:52< Crab_> Aethaeryn: yes
20100428 19:31:57< Aethaeryn> just a quick demo of what the person can do
20100428 19:32:03< Aethaeryn> no time to play a full game since I have class in an hour
20100428 19:32:19< Crab_> Aethaeryn: I guess it would be an easyCoding task to code a console command to switch between different games
20100428 19:32:44< Crab_> Aethaeryn: no, I can't look now (i'm at work for 30minutes or so, then leave for home)
20100428 19:33:08< Aethaeryn> ah, okay
20100428 19:33:18< Crab_> Aethaeryn: but the problem is with the UI
20100428 19:33:23< Aethaeryn> I won't be home for another 4 hours
20100428 19:33:29< Aethaeryn> but I meant something like a right click "switch map"
20100428 19:33:42< Aethaeryn> In that sense, it would be almost like it's the same scenario, just many simultaneous maps
20100428 19:33:57< Crab_> Aethaeryn: we can code a prototype and see how it goes.
20100428 19:34:22< Aethaeryn> basically the game in mind, p1 is a game master
20100428 19:34:23< Crab_> Aethaeryn: we'll need a way to signal 'hey, another layer needs attention' or 'hey, another layer has a chat mesasge'
20100428 19:34:35< Aethaeryn> who has a kind of "debug mode"
20100428 19:34:43< Aethaeryn> he can spawn items, change terrains, raise/lower walls.
20100428 19:34:46< Aethaeryn> and several other htings
20100428 19:34:47< Aethaeryn> *things
20100428 19:34:53< Crab_> Aethaeryn: oh, nice :)
20100428 19:34:59< Aethaeryn> So it's almost like Wesband where every game is different.
20100428 19:35:05< norbert_> I played a map that could do that a couple of days ago
20100428 19:35:17< Aethaeryn> But that's because the host on-the-fly maps and puts units
20100428 19:35:19< Crab_> Aethaeryn: then the issues will be less a problem
20100428 19:35:28< Aethaeryn> But I was going to add a stairwell
20100428 19:35:29< Ivanovic> Crab_: but, uhm, it will basically be something completely different from the current stats stuff
20100428 19:35:32< Aethaeryn> that basically puts you to a new blank map
20100428 19:35:48< Aethaeryn> and ideally, the host should be able to switch between maps and players could go on their own dungeon adventures or something
20100428 19:35:59< Aethaeryn> in reality, it'll probably just switch to a new map from the previous :P
20100428 19:36:05< norbert_> 5p_-_Master_of_Dungeons_Turn_109_(1986).gz
20100428 19:36:07< Crab_> Ivanovic: yes, not sure about 'completely', but different. at least the data will be used and some info will be displayed.
20100428 19:36:08< norbert_> that one
20100428 19:36:09< norbert_> :)
20100428 19:36:12< Aethaeryn> norbert_: Yeah. I made that.
20100428 19:36:17< norbert_> I know
20100428 19:36:28< Aethaeryn> Just did some revisions today.
20100428 19:36:33< Ivanovic> Crab_: uhm, no, the current data will not be used by what norbert_ wants to creat
20100428 19:36:38< Aethaeryn> To make it fully work with Bob's RPG Era
20100428 19:36:47< Crab_> Ivanovic: why ?
20100428 19:36:53< Aethaeryn> Now all the feature list are nice items, not necessary items.
20100428 19:36:57< Aethaeryn> *feature todo list
20100428 19:37:02< Ivanovic> if i understood it correctly his data is not about things like how singleplayer campaign scenarios are completed
20100428 19:37:18< Ivanovic> but more about things like "which maps are played most often" and the likes
20100428 19:37:38< Aethaeryn> norbert_: I'm going to host another one in about 4 hours
20100428 19:37:40< Ivanovic> quite contrary things
20100428 19:37:44< Aethaeryn> If all goes well.
20100428 19:37:47< Aethaeryn> With the new version.
20100428 19:37:55< Ivanovic> so: the data norbert_ wants to have can most likely be gatherd directly on the server side
20100428 19:38:00< Crab_> norbert_: what you'll say about this ? ^
20100428 19:38:08< Ivanovic> where our data was stuff that is not singleplayer related at all
20100428 19:38:13< norbert_> Crab_: that both things would be interesting to use/display
20100428 19:38:20< Ivanovic> so that at least the main menu entry can be completly be removed
20100428 19:38:55< Ivanovic> (and yeah, i think it is reasonable to gather data about (observable) games and make that stuff available)
20100428 19:38:58< norbert_> I can just start out with displaying regular numbers (the results of queries) and go from there, diagrams, bars, and whatnot
20100428 19:39:01-!- dtiger [~dtiger@dynamic-vpdn-93-125-16-74.telecom.by] has joined #wesnoth-dev
20100428 19:39:27< Ivanovic> stats like for example which maps are played most often and which player wins most often as well as eras used and the likes
20100428 19:39:40< norbert_> yep
20100428 19:40:12< Ivanovic> reasonable stats to gather right on the serverside, stats that could eg be extracted from the saved replays
20100428 19:40:20< norbert_> nice circle plot diagram of the eras that are being used
20100428 19:40:51< Crab_> Aethaeryn: ok, now I understand the idea..
20100428 19:41:28< Aethaeryn> Basically, I want the host player to be able to have multiple dungeon maps up at once.
20100428 19:41:28< Crab_> Aethaeryn: I think that it'll be possible to hack that in, but I need to think about the best way to do it
20100428 19:41:35< Aethaeryn> And each player could take their own path or whatever.
20100428 19:41:44< Aethaeryn> Yeah, it's not a big deal at all.
20100428 19:41:46< Aethaeryn> Just a "would be nice"
20100428 19:41:52< Aethaeryn> To have multiple interacting maps
20100428 19:42:32< CIA-10> caslav_ilic * r42294 /branches/1.8/po/ (64 files in 16 dirs): Updated translations for Serbian (intentionally not applying to trunk too).
20100428 19:43:21< Ivanovic> Crab_: so even with the stats gathering stuff norbert_ wants to have the current system can be removed
20100428 19:43:43< Ivanovic> the system that once upon the time was meant to help campaign creators balance their stuff
20100428 19:43:59< norbert_> how is that?
20100428 19:44:11< Ivanovic> but at the beginning was not scalling well (vector graphs that took ages to generate) and thus the pages were damn slow
20100428 19:44:12< norbert_> the code is still required for local campaign games
20100428 19:44:23< Ivanovic> and some time after it was completely unusable
20100428 19:44:37-!- ancestral [~ancestral@97-116-122-90.mpls.qwest.net] has joined #wesnoth-dev
20100428 19:44:41< norbert_> or were you referring to MP code
20100428 19:44:43< Ivanovic> that is: for something like >2years you could not really use the system, i'd even say 3 years
20100428 19:45:55< Crab_> Ivanovic: yes, if we remove the actual stats collection routines, then it'll be not possible to get info about SP games.
20100428 19:46:14< Crab_> Ivanovic: you're right about the 'link to see my stats'
20100428 19:46:50< Ivanovic> the available system was basically not working at all for a long while
20100428 19:47:16< Ivanovic> at least it was not usable for content creators and the "view my data" stuff was, uhm, not working well either
20100428 19:47:23< boucman> gabba: I have to leave for the evening, I'll try to join before bedtime, hopefully you will still be there... if not, let's try to communicate via the dev-ml until we meet on IRC (dev-ml is the only wesnoth communication mean I have at work)
20100428 19:47:38< Ivanovic> norbert_: the question is: what would you gather regarding singleplayer stats?
20100428 19:48:03< norbert_> Ivanovic: it's already being gathered
20100428 19:48:06< Ivanovic> norbert_: consider this scenario: there is *no* other user of singleplayer data
20100428 19:48:09< Ivanovic> norbert_: NO, it is not
20100428 19:48:10< norbert_> what is there will do fine, for now
20100428 19:48:27< Ivanovic> there is NOTHING running anymore on our main server because it is BROKEN!!!
20100428 19:48:35< Ivanovic> no collection routines running anymore, no data collected!
20100428 19:48:44< norbert_> so Crab_ was wrong then
20100428 19:48:49< Ivanovic> we have no more stats available
20100428 19:48:57< Crab_> might be, but I think that the collection routines are easy enough to fix.
20100428 19:49:10< Ivanovic> yeah, there is some code that could upload things, but we have stopped the sinks because this stuff f***ed our server up!
20100428 19:49:32< Ivanovic> norbert_: so if you really want data the question remains: what do you really want to have?
20100428 19:49:46< Ivanovic> and no, a case of "i will use what is there" does not work
20100428 19:49:52< Crab_> then we need to scale down the data which is sent, to the minimum required
20100428 19:49:57< norbert_> it's not really important what I (personally) would like to have
20100428 19:49:58< Ivanovic> since what is there at the moment (on our server) is nothing
20100428 19:50:18< Ivanovic> it is important what you want to do with the data and what data you need for this
20100428 19:50:20< norbert_> what matters is what players and campaign developers would like to see
20100428 19:50:59-!- wesbot changed the topic of #wesnoth-dev to: 99 bugs, 273 feature requests, 16 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100428 19:51:06< norbert_> probably things like how many times they are played, how many turns and time to finish, what difficulties, how many deaths, and so on
20100428 19:51:32< norbert_> how people die and where on the map
20100428 19:51:37< CIA-10> ivanovic * r42295 /trunk/po/wesnoth-low/de.po: updated German translation
20100428 19:51:42< CIA-10> ivanovic * r42296 /branches/1.8/po/wesnoth-low/de.po: updated German translation
20100428 19:52:52-!- billynux [~c8078d05@gateway/web/freenode/x-nmqfarerbckrgkyp] has joined #wesnoth-dev
20100428 19:53:37< Crab_> hi, billynux
20100428 19:53:40< billynux> hi Crab_ 
20100428 19:53:58< billynux> and hi boucman, Ivanovic, fendrin, etc... :)
20100428 19:54:09< fendrin> hi billynux
20100428 19:54:33< Ivanovic> norbert_: since what you say is basically a case of "i'd like to have all data possible" it would be easiest to upload complete replays
20100428 19:54:39< Crab_> there's an ongoing discussion about the stats code - norbert_ is interested in making use of some SP stats, and we need to decide what to do with the existing stats collection code.
20100428 19:54:43< Ivanovic> the problem with this being how much you got to upload
20100428 19:55:02< Ivanovic> personally i'd prefer you work on generating something out of the existing replays
20100428 19:55:02-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100428 19:55:02< Crab_> 'entire replays' is very large and includes a lot of useless stuff...
20100428 19:55:26< Ivanovic> since we got *lots* of them from the mp server and there will probably be lots of information avainale from them
20100428 19:55:50< billynux> ok Crab_, is it the same "stats" stuff I'm suppose to remove? (I'll start to look into it tomorrow)
20100428 19:56:53< Ivanovic> Crab_: collecting all "usefull data" is lots of work, too
20100428 19:57:06< Crab_> billynux: yes. if norbert_ will use only the MP stuff, he won't need sp stats uploader and the feature can be removed. if norbert_ will volunteer to use (in a good way) a minimal amount of SP stats, then we'll need to retain it.
20100428 19:57:12< Ivanovic> and honestly i don't trust too much in what we collect at the moment
20100428 19:57:18-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100428 19:57:35< Crab_> Ivanovic: yes, I understand
20100428 19:57:57< Crab_> norbert_: so, let's figure out the SP stats we'll collect for now.
20100428 19:58:13< billynux> Ok then. I await decision on the subject...
20100428 19:58:21-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: class...]
20100428 19:58:38< norbert_> a minimal amount of SP stats could be generated by logging some simple things for now, 1. when someone dies or when someone finishes 2. a scenario of the whole 3. campaign
20100428 19:59:00< Crab_> norbert_ : what do you mean by 'dies' ? a 'defeat' ?
20100428 19:59:02< norbert_> at these moments, write away the current turn, current gold, number of units, time played, maybe some other things
20100428 19:59:13< norbert_> if the player dies
20100428 19:59:16< Crab_> norbert_: so, '1 row for scenario' is enough ?
20100428 19:59:21-!- timotei [~timotei@193.34.191.4] has quit [Read error: Connection reset by peer]
20100428 19:59:31< norbert_> no, a couple of rows per scenario
20100428 19:59:39< Crab_> why couple ?
20100428 19:59:40-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100428 19:59:43< norbert_> 1 row if someone just finishes it
20100428 19:59:51< norbert_> but more if they die 1-n times
20100428 20:00:01< Crab_> well, that's N games...
20100428 20:00:04< Crab_> and 1 row per game
20100428 20:00:26< Crab_> so, to clarify "1 row per game" ?
20100428 20:01:22< norbert_> campaign | scenario | event (end or die) | gold | turn | time since x | ...
20100428 20:01:25< norbert_> something like that?
20100428 20:01:35< Crab_> yes
20100428 20:01:48< Crab_> based on http://wiki.wesnoth.org/SoC_Ideas_Stats_Server#Database_Details, the GAMES table there.
20100428 20:01:50< norbert_> that would be very usable
20100428 20:01:51-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100428 20:02:03< mordante> servus
20100428 20:02:14< Crab_> norbert_: can you give a rough time estimate for the 'when you'll be able to use this info?'
20100428 20:02:31< Crab_> norbert_: and what is required from us ?
20100428 20:02:44< Crab_> mordante: hi, mordante
20100428 20:02:52< mordante> hi Crab_ 
20100428 20:03:15< norbert_> sorry, I cannot
20100428 20:03:35< norbert_> I could try that, I've worked on some IT projects, and I always fail at giving estimates :)
20100428 20:03:40< Crab_> norbert_: ok :)
20100428 20:03:54< Crab_> norbert_: how much time you've needed for your website ?
20100428 20:04:07< billynux> hi mordante 
20100428 20:04:17< Crab_> mordante: there's an ongoing discussion about the stats code - norbert_ is interested in making good use of some SP stats, and we need to decide what to do with the existing stats collection code.
20100428 20:04:19< norbert_> coding was only a fraction of the time spend on that
20100428 20:04:25< mordante> hi billynux 
20100428 20:04:45< mordante> Crab_, ok will read it in the logs
20100428 20:04:59< Crab_> norbert_: and what about the second question ? what do you need from us ? a daily portion of sql dump via scp?
20100428 20:05:17< mordante> Crab_, are we keeping them?
20100428 20:05:29< norbert_> Crab_: again, it depends on how much MB this would be per day...
20100428 20:05:53< norbert_> Crab_: if it's under, say, 5 MB, I'd prefer to receive it by e-mail
20100428 20:06:09< norbert_> and daily, yes
20100428 20:06:12-!- qemqemqem [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has quit [Ping timeout: 276 seconds]
20100428 20:06:16< Crab_> mordante: Ivanovic says that we're not.
20100428 20:06:35< Crab_> mordante: but we can fix it
20100428 20:07:35< Crab_> norbert_: so, then the plan is 1) scale down the c++ side of stats collection, don't collect all unnecessary data, just the bare minimum.
20100428 20:08:01< norbert_> yes, so the server can handle it
20100428 20:08:14< Crab_> 2) fix the collector script to actually store the data (even 'to files')
20100428 20:08:45< Crab_> 3) hack up a script to lazily preprocess the data, and store them to db
20100428 20:09:15< Crab_> 4) hack up a script to prepare a archive with the (anonymized) daily portion of the data.
20100428 20:09:37< Crab_> 5) decide on transport, and hack up a script to transport the data to you
20100428 20:10:07< Crab_> 6) you'll make use of the data on your side to display info about the popularity of sp scenarios (and other data, later)
20100428 20:10:14< Crab_> norbert_: is it good ?
20100428 20:10:38< norbert_> yes, 2) and 3) maybe in one go
20100428 20:10:55< norbert_> same for 4) and 5)
20100428 20:11:24< norbert_> but that'll work
20100428 20:11:44< billynux> I gotta go. I'll log in tomorrow and I'll be here most of the day. I can look at the logs to see whats your ruling on the stats code, bye Crab_ mordante ...
20100428 20:11:51< Crab_> billynux: bye
20100428 20:11:56< mordante> ok bye billynux 
20100428 20:11:57-!- billynux [~c8078d05@gateway/web/freenode/x-nmqfarerbckrgkyp] has quit [Quit: Page closed]
20100428 20:12:26< Crab_> ok. mordante, Ivanovic ^ what do you think about reanimating the SP stats, in that scaled-down way ?
20100428 20:12:52< Ivanovic> personally i'd prefer to see some stats system working first that is based upon existing replays
20100428 20:13:35< mordante> Crab_, I finish the logs first
20100428 20:13:51< Ivanovic> using 100% freely available data (as in: the stuff we offer for everybody from our server)
20100428 20:14:08< timotei> wesbot: seen YoghiHH ?
20100428 20:14:08< wesbot> timotei: Sorry, I don't know of YoghiHH.
20100428 20:14:13< timotei> wesbot: seen YogiHH ?
20100428 20:14:14< wesbot> timotei: The person with the nick YogiHH last spoke 20h 40m ago.  19h 41m ago they left with the message: Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]
20100428 20:14:28< Crab_> Ivanovic: I expect that many will switch from 1.8 to 1.8.1 because of MP lobby issues, so, we need to decide about 'what to do about SP stats in 1.8' now - should we remove it fully or remove half of it ?
20100428 20:14:35< timotei> mordante, If I want to use git instead of svn for wesnoth, I use the same link for: ssh+svn:// etc
20100428 20:14:45< timotei> mordante, or I need to use a special, git link?
20100428 20:14:49< Ivanovic> Crab_: for 1.8: remove fully
20100428 20:14:50< Crab_> timotei: the same link
20100428 20:14:57< timotei> ok thanks Crab_ 
20100428 20:15:07< mordante> timotei, same link best download the git tarball we hide on the server
20100428 20:15:15< mordante> then you only need to modify a config file
20100428 20:15:30< timotei> why not checkout with: git svn init svn+ssh:// ?
20100428 20:15:47< Crab_> timotei: it will fetch each and every release
20100428 20:15:49< timotei> the tarball may be outdated?
20100428 20:15:56< norbert_> Ivanovic: is the code in a really bad shape?
20100428 20:15:59< timotei> I will fetch only HEAD :P
20100428 20:16:04< Crab_> timotei: it's better to start with a snapshot and fetch the update
20100428 20:16:15< Crab_> timotei: well, local history is nice to have
20100428 20:16:18< Ivanovic> norbert_: honestly i do not trust the code we got for gathering data
20100428 20:16:18< mordante> timotei, you have a few days ;-)
20100428 20:16:38< Crab_> norbert_: are you ok with the fact that 1.8 will not upload any data, but the data will come in only from 1.9 clients (so, it will be several months before any real amount of data will come in)
20100428 20:16:52< norbert_> Crab_: sure
20100428 20:16:54< Crab_> norbert_: the plus side is that there will be no time constraints to code everything right.
20100428 20:17:03< Ivanovic> norbert_: that is why i would love to see a system based on replays first
20100428 20:17:03< timotei> well, you lost me. the tarball, isn't already the last version? so it would be the same thing of fetching head, isn't it?
20100428 20:17:03< mordante> timotei, if you download it manually it takes a long time, better get the tarball and only update the last changes
20100428 20:17:06< norbert_> that's good, no time constraints
20100428 20:17:11< timotei> oh
20100428 20:17:12< timotei> ok:)
20100428 20:17:13< Ivanovic> since for the mp server we got thousands of replays available directly
20100428 20:17:20< Ivanovic> so you can use the stuff right away
20100428 20:17:29< mordante> timotei, no the tarball is created daily (I think) so is slightly behind
20100428 20:17:32< Crab_> Ivanovic: ok, good. so, let's "remove for 1.8, start lazily fixing it for 1.9" ?
20100428 20:17:32< norbert_> Ivanovic: it's different though, because I have to parse the gz files myself
20100428 20:17:42< timotei> mordante, where I get the tarball from?
20100428 20:17:46< norbert_> Ivanovic: it's way easier with db data
20100428 20:17:53< timotei> cause I want to move on git, svn makes too much overhead
20100428 20:17:57< mordante> timotei, after downloading you need to modify .git/config to use the right nick
20100428 20:18:00< timotei> 3.2gigas is my wesnoth folder :D
20100428 20:18:04< timotei> yep
20100428 20:18:37< mordante> timotei, http://www.wesnoth.org/files/hidden/wesnoth-git.tar.bz2
20100428 20:19:27< mordante> timotei, and when you use git you need to run git gc on a regular base, which temporary takes quite a bit of disk space
20100428 20:19:58< mordante> normal git is rather efficient, but downloading with svn gives a lot of garbage
20100428 20:20:10< mordante> best also turn off the auto gc feature
20100428 20:20:27< timotei> wow, 1.7 gigas :D
20100428 20:20:36< timotei> what's so big?:|
20100428 20:20:38< mordante> timotei, it's the entire history
20100428 20:20:44< norbert_> Crab_: for 1.9 sounds fine; I'll work on other things for now, like the MP stats, the lobby bot and RFC, that authors can update data, and so on
20100428 20:20:49< timotei> ok
20100428 20:21:07< timotei> I understand now, so even if no internet connection, I can take a look at different versions?:)
20100428 20:21:18< mordante> yes
20100428 20:21:23< timotei> cool
20100428 20:21:35< mordante> and even commit things and 'upload' them when you have a connection again
20100428 20:21:45< timotei> yeah, that I know
20100428 20:21:52< timotei> so: git gc on the directory of wesnoth, let's say daily?
20100428 20:22:14< mordante> git count-objects tells you how much `garbage' you collected
20100428 20:22:41< mordante> just depends on the updates, pot-updates create a lot of `garbage'
20100428 20:22:54< timotei> what's pot-updates?
20100428 20:22:57< norbert_> have to go, talk to you all later
20100428 20:22:58-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has quit [Quit: Leaving]
20100428 20:23:05< mordante> the ones Ivanovic commits
20100428 20:23:28< mordante> Crab_, Ivanovic I've no strong opinion about the stats, but maybe first indeed do the MP stats and think about SP stats later
20100428 20:23:51< timotei> Ivanovic, what's a "pot-update"?
20100428 20:23:54< mordante> I prefer to see that it works and for a longer time
20100428 20:23:58< timotei> or a, "pot"
20100428 20:24:11< mordante> timotei, update of the base translation files
20100428 20:24:11< Ivanovic> those updates are for the translation files/catalogs
20100428 20:24:30< timotei> oh
20100428 20:24:33-!- qemqemqem__ [~quassel@urwireless-dhcp-128-151-25-215.wireless.rochester.edu] has joined #wesnoth-dev
20100428 20:24:33< mordante> and these commits often modify a lot of files and are huge
20100428 20:24:33< timotei> ok
20100428 20:24:52< Ivanovic> a size of several MB is not uncommon for those updates
20100428 20:25:03< timotei> why are they so big?
20100428 20:25:27< timotei> is just "text" like code, isn't it?
20100428 20:25:38< Ivanovic> yes, but there are 50 translations
20100428 20:25:47< Ivanovic> and each file for each translation is update
20100428 20:25:48< Ivanovic> s
20100428 20:25:51< Ivanovic> s/s/d
20100428 20:25:54< Crab_> mordante: well, then we remove the stats fully for 1.8. but, I would like to try to fix things for 1.9, so I think it will be a good idea to not remove the actual 'stat submission' code from trunk. it should be quite fast to fix, since nothing to add, only remove unnecessary stuff.
20100428 20:26:42< mordante> Crab_, sounds like a plan then we can decide before the 1.10 release what to do with the stuff in trunk
20100428 20:26:46< Crab_> ok
20100428 20:26:55< mordante> either norbert has is working or not and we remove it
20100428 20:26:59< Crab_> yes
20100428 20:27:07< Crab_> thanks / bye
20100428 20:27:10-!- Crab_ [~Crab@wesnoth/developer/crab] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20100428 20:27:30< mordante> Ivanovic, you also agree with that?^
20100428 20:27:54< Ivanovic> sounds good
20100428 20:28:01< loonycyborg> Removing stats from 1.8 doesn't seem optimal to me considering that most clients will be this version for some time..
20100428 20:28:16< Ivanovic> loonycyborg: and we got nothing really using any of the incoming data
20100428 20:28:38< Ivanovic> loonycyborg: so why should we leave it there and have the server send 404 messages?
20100428 20:30:09< loonycyborg> But if it's planned to revive it that's only a temporary issue..
20100428 20:31:09-!- timotei21 [timotei@188.24.7.149] has joined #wesnoth-dev
20100428 20:31:53< Ivanovic> loonycyborg: we got no idea what we will revive in which way
20100428 20:32:07-!- gabba [~gabriel@70.35.167.54] has joined #wesnoth-dev
20100428 20:32:16< gabba> hi
20100428 20:32:50< loonycyborg> That is if it can't use data from 1.8 then removing the client functionality does make sense, but if it's planned to revive it then ignoring 1.8 seems suboptimal.
20100428 20:32:58< timotei21> hi gabba 
20100428 20:33:01-!- timotei21 is now known as timotei_away
20100428 20:33:11-!- timotei_away is now known as timotei21
20100428 20:33:18< gabba> hi timotei21
20100428 20:34:03-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has joined #wesnoth-dev
20100428 20:34:23-!- timotei [~timotei@193.34.191.4] has quit [Ping timeout: 246 seconds]
20100428 20:35:10< Ivanovic> loonycyborg: honestly i am not perfectly fine with sharing a database dump of the data that we currently collect
20100428 20:35:31< Ivanovic> since for the players there is no way to know where we send the data to that we currently do connect
20100428 20:35:39< Ivanovic> beside this the info we provide is plain wrong
20100428 20:36:05< Ivanovic> it is *not* possible to view your statistics at the link we provide in the dialog and most likely will never be possible
20100428 20:36:33< Ivanovic> this alone are reasons for me to remove it at least from 1.8
20100428 20:37:49< loonycyborg> Indeed. That feature doesn't look very privacy-friendly to me too.
20100428 20:38:12< loonycyborg> But that would be argument for removing it altogether :P
20100428 20:38:20< Ivanovic> yes
20100428 20:38:39< Ivanovic> what i am fine with is analysing those games that were visble in public
20100428 20:39:02< Ivanovic> so games played on the official server with observers being allowed
20100428 20:39:13-!- mordante_ [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100428 20:39:29< Ivanovic> since everybody has access to those replays it is fine if they are used to gather some data
20100428 20:41:21< gabba> boucman: I'll be around for a while, just ping me when you come back
20100428 20:41:39< gabba> boucman: if we miss each other, I'll answer on the mailing list
20100428 20:42:30-!- timotei21 is now known as timotei
20100428 20:42:30-!- timotei is now known as timotei_away
20100428 20:42:41-!- timotei_away is now known as timotei
20100428 20:42:57-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Ping timeout: 268 seconds]
20100428 21:10:55-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Quit: This computer has gone to sleep]
20100428 21:14:13-!- drusepth [~drusepth@dhcp-077-249-151-209.chello.nl] has joined #wesnoth-dev
20100428 21:20:02< gabba> Ivanovic: I just tried, and I still can't update my proposal's abstract. Dunno how comes timotei could and I can't... There was no message on the announce list for GSoC students, either.
20100428 21:20:19< timotei> lol
20100428 21:20:26< timotei> well, let me recall what I did
20100428 21:20:39< gabba> how much did you bribe them timotei ? :P
20100428 21:20:58< timotei> well, you need to know what path to take for editing the proposal :))
20100428 21:21:14< timotei> you need to edit the public proposal
20100428 21:21:19-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100428 21:21:54< timotei> so, go on the list of Student PROJECTS
20100428 21:21:58< timotei> http://socghop.appspot.com/gsoc/student/list_projects/google/gsoc2010
20100428 21:22:05< timotei> not the list of proposal
20100428 21:22:15< timotei> click on that project, and voilla, you can update it:D
20100428 21:22:37< gabba> Ahhh, I see
20100428 21:22:41< timotei> :P
20100428 21:23:02< gabba> Seriously their "Melange" web app sucks 
20100428 21:23:07< timotei> :))
20100428 21:23:16< timotei> well, is just not so "user-friendly"
20100428 21:23:32< gabba> It's slow, and there's 5 different ways to get to the same info, but to do slightly different things to it
20100428 21:23:58< gabba> I'm surprised since google stuff is usually top-notch
20100428 21:24:21 * gabba stops ranting and goes to edit his abstract
20100428 21:24:22-!- deekay [~dk@wesnoth/developer/dragonking] has quit [Quit: deekay]
20100428 21:24:26< timotei> maybe melange was did during a gsoc project
20100428 21:24:27< timotei> :))
20100428 21:24:33< gabba> thanks timotei for the tip
20100428 21:24:37< timotei> np
20100428 21:24:40< mordante_> gabba, they like bug reports ;-)
20100428 21:24:58< timotei> maybe Ivanovic should tell the others the way to modify that
20100428 21:25:03< gabba> mordante_: does "your app generally sucks" count as a bug report? :P
20100428 21:25:20< timotei> maybe more of a "non-feature request"
20100428 21:25:21< timotei> =))
20100428 21:25:25< mordante_> gabba, no but if you have suggestions how to improve they like it
20100428 21:25:35< mordante_> and melange is no google project ;-)
20100428 21:25:40< timotei> :P
20100428 21:25:54< timotei> I would like an interface like a "tree"
20100428 21:25:59-!- qemqemqem__ [~quassel@urwireless-dhcp-128-151-25-215.wireless.rochester.edu] has quit [Ping timeout: 260 seconds]
20100428 21:26:06< gabba> mordante_: I'll keep that in mind, hopefully they don't think bug reports like Ubuntu heh
20100428 21:26:07< timotei> n-ary tree
20100428 21:26:40< gabba> s/think/treat
20100428 21:26:41< mordante_> I've no idea what you mean about Ubuntu bugs
20100428 21:27:05< mordante_> ah now I think I understand
20100428 21:27:40< gabba> mordante_: well, on the Ubuntu tracker bugs often get ignored for a loooong while, and then you either get an automated message or they get automatically deleted, no matter how good your report is or how much effort you put into it
20100428 21:28:18< CIA-10> mordante * r42297 /trunk/src/unit_frame.hpp: 
20100428 21:28:18< CIA-10> Make a member function a const member function.
20100428 21:28:18< CIA-10> Issue found by cppcheck.
20100428 21:28:22< CIA-10> mordante * r42298 /trunk/src/game.cpp: Initialize all members.
20100428 21:28:27< mordante_> gabba, ah that's indeed annoying
20100428 21:28:29< gabba> mordante_: so in short reporting bugs there is often a waste of time, unfortunately
20100428 21:29:00< mordante_> well they get their `reward' in the end 0 bugs
20100428 21:29:33< gabba> mordante_: exactly: "0 bugs" (rofl)
20100428 21:29:58< shadowmaster> I have just noticed that our bug counter is at 99
20100428 21:30:04< shadowmaster> that's nice
20100428 21:30:40< mordante_> timotei, I think r42298 is your doing ;-) it's more efficient to initialize items in the initializer list else they get default constructed and initialized in the body afterwards
20100428 21:30:44< shadowmaster> s/(nice)/un$1/
20100428 21:31:26< mordante_> gabba, yup
20100428 21:31:42-!- dtiger [~dtiger@dynamic-vpdn-93-125-16-74.telecom.by] has quit [Remote host closed the connection]
20100428 21:32:00-!- Zarel [~Zarel@warzone2100/developer/Zarel] has joined #wesnoth-dev
20100428 21:32:05< gabba> mordante_: I had a concrete experience of that recently
20100428 21:33:33< gabba> one guy kept moving my bug to the fglrx package (i.e. the dumpbin) even though it was asking for a way for normal users to escape a frozen full-screen app
20100428 21:33:50< timotei> mordante_, I don't understand
20100428 21:33:56< timotei> what's with r42298?
20100428 21:34:09< shadowmaster> gabba: ew, fglrx
20100428 21:34:10< timotei> ah, ok, let me check
20100428 21:34:14< mordante_> short for revision 42298 <CIA-10> mordante * r42298 /trunk/src/game.cpp: Initialize all members.
20100428 21:34:15 * shadowmaster shudders
20100428 21:34:44< gabba> the only link with fglrx was that I have an ati card, but the freeze I was mentioning as an example was Wesnoth with pulseaudio :P :D
20100428 21:35:03< gabba> It would be hilarious if it wasn't sad
20100428 21:35:09 * mordante_ shudders Ubuntu and their pulseaudio
20100428 21:35:18-!- mordante_ is now known as mordante
20100428 21:35:19< timotei> mordante_, oh, I understand
20100428 21:35:28< timotei> thanks for the tip
20100428 21:35:29< shadowmaster> someone should kill PA and bury it deep in the earth
20100428 21:35:40< mordante> ok good you're welcome
20100428 21:35:55< shadowmaster> somewhere were nobody would accidentally bring it back to the surface. Antarctica maybe
20100428 21:35:59< shadowmaster> *where
20100428 21:36:17-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100428 21:37:17< gabba> shadowmaster: the concept of PA is good, and I like how you can change individual applications' volume, but the execution is horrible
20100428 21:37:48< shadowmaster> I think it shouldn't be the default, at least.
20100428 21:38:20< Soliton> gabba: can't you switch to a tty still?
20100428 21:38:21< loonycyborg> I heard that only its Ubuntu deployment is horrible.
20100428 21:38:22< shadowmaster> I personally don't see the need for an(other) abstraction layer when most apps deal with ALSA (through abstraction libs) just fine
20100428 21:38:52< shadowmaster> unless, I run a sound studio, have very powerful and Cool™ sound hardware, or something
20100428 21:38:58< gabba> shadowmaster: I agree... or rather Ubuntu should have a permanent BETA label, instead of bragging all the time about how good they are
20100428 21:39:04< shadowmaster> :P
20100428 21:39:24< shadowmaster> well, thankfully Debian still sticks to ALSA AFAIK
20100428 21:39:28< gabba> Soliton: yes, I can kill it from the console
20100428 21:39:51< shadowmaster> at least they didn't slap the PulseAudio packages on my face when I switched to Testing now...
20100428 21:40:02< Crab_> gabba: reading through your email to wesnoth-dev ml
20100428 21:40:04< Soliton> gabba: not good for normal users? :-)
20100428 21:40:57< gabba> Soliton: exactly... in past years I tried to switch pretty normal Windows users to Ubuntu, and it was a disaster, because of unpolished stuff like that
20100428 21:41:02< timotei> tbh, I updated my linux mint yesterday, and it updates pulseaudio too. After that: no sound. I uninstalled it=> sound is back again
20100428 21:41:21< gabba> Crab_: ok :)
20100428 21:41:30< timotei> both linux and windows have their issues
20100428 21:41:34< timotei> what about solaris?:D
20100428 21:42:48< gabba> shadowmaster: is the debian desktop usable these days, and do you have to jump through hoops to have all desktop apps (and games if possible) up-to-date?
20100428 21:43:01< Crab_> gabba: take a look at src/ai/actions.cpp  - you'd be able to spot a few things like 'recall action is needed, similar to the recruit one' and 'attack has a parameter - weapon to use'
20100428 21:43:10< shadowmaster> gabba: um, I use KDE 4
20100428 21:43:24< shadowmaster> usable enough for my taste and I can bear the occasional plasma-desktop or kwin crashes
20100428 21:43:27< Soliton> depends what you mean with up-to-date.
20100428 21:43:59< CIA-10> mordante * r42299 /trunk/data/gui/default/ (widget/button_default.cfg window/title_screen.cfg): 
20100428 21:43:59< CIA-10> Readd the upload stastics to gui2.
20100428 21:43:59< CIA-10> The new plan is to only revert for the 1.8 branch and see whether it can be
20100428 21:43:59< CIA-10> fixed in trunk. (Reverts r42287).
20100428 21:44:04< gabba> By up-to-date I mean upgrading to the latest stable firefox or openoffice as soon as they get out... they usually have major improvements
20100428 21:44:05< CIA-10> mordante * r42300 /trunk/src/gui/ (2 files in 2 dirs): 
20100428 21:44:05< CIA-10> Enable some error messages.
20100428 21:44:05< CIA-10> They were blocked by the stringfreeze before.
20100428 21:44:11< shadowmaster> it currently doesn't crash unless I switch compositing on/off too often
20100428 21:44:18< Soliton> distributions can't ever be as up-to-date as downloading an installer from some website.
20100428 21:44:28< shadowmaster> also, the latest stable Firefox was 2.something when 3.0 was the "latest Firefox" :)
20100428 21:44:40< shadowmaster> (not talking about Debian in particular)
20100428 21:45:07< Crab_> mordante: thanks
20100428 21:45:13< gabba> Crab_: looking
20100428 21:45:24< Soliton> i agree that removing stats uploading from 1.8.1 is silly, btw.
20100428 21:45:37-!- Sirp__ [~97c1dc1b@wesnoth/developer/dave] has joined #wesnoth-dev
20100428 21:45:46< Soliton> i don't see what needs to be rethought about what stats we want to collect.
20100428 21:45:57< mordante> Crab_, well I hope to get the title screen ready for 1.10 ;-)
20100428 21:46:11< Crab_> mordante: what blocks it :) ?
20100428 21:46:43< Crab_> gabba: also, I think that you should make 'conflicts' a separate data structure
20100428 21:47:06< mordante> Crab_, it looks ugly on small screens and doesn't show the translated pictures properly
20100428 21:47:20< gabba> Soliton: I dislike installers under linux since there's no central place that tracks them, and I forget what I have installed after a few months. Ubuntu is so popular that you can usually find .debs for the latest software, and that's very good.
20100428 21:47:34< mordante> Crab_, and I lack time to fix it, but with gsoc really running I expect to have more time
20100428 21:47:52< mordante> no more reading huge amounts of proposals ;-)
20100428 21:47:58< mordante> of*
20100428 21:48:03< Crab_> gabba: also note that recruit/recall has a 'acting unit' - it's the leader which recruits/recalls
20100428 21:48:15< Crab_> gabba: this might get important if we add per-leader recruit lists
20100428 21:48:37< Soliton> gabba: you either download installers or shady .debs or whatever yourself or you wait until your distribution updates. there isn't really another way.
20100428 21:49:24-!- shadowmaster is now known as frankenstein
20100428 21:50:00-!- frankenstein is now known as shadowmaster
20100428 21:50:12< mordante> I'm off night
20100428 21:50:23< gabba> bye mordante
20100428 21:50:40-!- Sirp__ [~97c1dc1b@wesnoth/developer/dave] has left #wesnoth-dev []
20100428 21:50:46-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100428 21:51:07< gabba> Crab_: "make 'conflicts' a separate data structure": you mean entirely parallel to planned actions, so I'd have to iterate separately on it?
20100428 21:53:38-!- Johannes13_ is now known as Johannes13
20100428 21:53:47< Soliton> isn't a conflict just a planned action with some status bit set?
20100428 21:54:44< gabba> timotei: "both linux and windows have their issues": of course, but trust me, from my tests on many innocent users I can affirm that windows is tested way better than any currently available linux-based desktop.
20100428 21:54:58< gabba> timotei: and I find it's a sad state of things
20100428 21:55:03-!- qemqemqem__ [~quassel@cpe-74-74-155-246.rochester.res.rr.com] has joined #wesnoth-dev
20100428 21:55:45-!- Sirp__ [~97c1dc1b@wesnoth/developer/dave] has joined #wesnoth-dev
20100428 21:56:56< Crab_> gabba: yes
20100428 21:57:11< timotei> unfortunately yes...
20100428 21:58:03< Crab_> Soliton: well, 'conflict' is a thing that makes little sense for a single planned action, and is only possible when multiple actions are considered. you're right about the possibility of the 'conflict bit' as a cache.
20100428 21:58:35< Crab_> Soliton: but yes, sorry, there can be 1-action conflicts, too.
20100428 21:58:43< gabba> Crab_: I'm not sure it's a good idea: for instance I want to draw the arrow differently if there's a conflict; if the conflicts are kept inside each planned move, I just have to check "is your conflict list empty? if so, draw the arrow normally"
20100428 21:59:01< Soliton> at least a conflicting action is still very much like a normal planned action, right?
20100428 21:59:16< Crab_> Soliton: yes, the action itself is the same
20100428 21:59:17< Soliton> they might get put in a different list though or so.
20100428 22:00:51< Crab_> gabba: yes, you can. what will be the plan for recalculating conflicts ?
20100428 22:01:23< Crab_> hard conflicts are easy (foreach action, translate to ai action, check if it's valid)
20100428 22:01:47< Crab_> but what about soft conflicts ?
20100428 22:02:29< gabba> Crab_: soft conflicts: for each pair of actions, check if they target the same hex, if so mark both as conflicted on that hex
20100428 22:02:50< Crab_> zookeeper: around ?
20100428 22:03:15< gabba> Crab_: maybe I could sort them by hex first to make this more efficient, dunno what it implies
20100428 22:03:56< Crab_> gabba: note that 'target' means 'end up on this turn'
20100428 22:04:14< Crab_> gabba: e.g., if you have two 'long movement' actions, they might conflict if they have the same next_hop
20100428 22:04:29< gabba> Crab_: true
20100428 22:04:29< Crab_> gabba: or consider the 'teleport' case
20100428 22:04:51< Crab_> gabba: can we block the teleport for a friend?
20100428 22:05:00< gabba> Crab_: and you're making me think I probably need to store the whole path or at least every hop
20100428 22:05:29< Crab_> gabba: whole path = every hop
20100428 22:05:57< Crab_> gabba: 'teleport' is just a normal movement in that regard
20100428 22:06:18< Crab_> gabba: and note that [teleport] can be (potentially) not only on villages  but pretty much anywhere
20100428 22:06:46< gabba> Crab_: each hop is the next hex in the path?
20100428 22:07:05< Crab_> yes. plus the start hex
20100428 22:07:24< Crab_> you can store only the part which is actually reachable this turn
20100428 22:08:01< Crab_> and be sure to note that the path might be recalculated if a move is done. e.g. , the action A might conflict with B or C depending on whether D is done.
20100428 22:08:06-!- Ken_Oh [~briang@static-71-178-174-220.washdc.fios.verizon.net] has quit [Read error: Connection reset by peer]
20100428 22:08:38< Crab_> e.g., "if we A want to teleport and that unit D moves out of the way, we'll conflict with B, else we'll conflict with C
20100428 22:09:08< gabba> Crab_: well, overall I just take into account the destination points of units for this turn, ignoring any possible future movement
20100428 22:09:26< gabba> Crab_: I don't think I need to worry about anything else
20100428 22:09:56< Crab_> yet 'for each pair of actions, check if they conflict' won't work reliably in all cases.
20100428 22:10:03< gabba> Crab_: then teleport isn't an extra worry, I just want to know where the player wants to send his unit
20100428 22:10:18< zookeeper> Crab_, yeah, briefly...
20100428 22:10:44< Soliton> simulate one action after the other and check for conflicts?
20100428 22:10:50< Soliton> they have an order, no?
20100428 22:10:59< Crab_> zookeeper: is there WML to allow the unit to modify the ZoC status of nearby units ?
20100428 22:11:33< Soliton> there is a distract ability, i think.
20100428 22:11:40< Crab_> Soliton: what it does ?
20100428 22:11:42< gabba> Crab_: I don't see how "the path might be recalculated if a move is done" will affect me, since the endpoint will either be still valid (so I don't care *how* you get there), or invalid, which means a hard conflict then
20100428 22:11:46< zookeeper> Crab_, well, sure WML can toggle any unit's ZoC on and off, but do you mean "automatically" like an ability?
20100428 22:11:52< Crab_> zookeeper: yes
20100428 22:11:55< zookeeper> distract only gives skirmisher to adjacent allies
20100428 22:12:06< Soliton> indeed.
20100428 22:12:08< zookeeper> there's no current way to make it work so that it'd actually affect the ZoC of enemies
20100428 22:12:08< Crab_> gabba: see ^
20100428 22:12:32< Crab_> zookeeper: ok, 'gives skirmisher to adjacent allies' - how it works ? do they lose the ability as soon as they start to move away ?
20100428 22:12:41< Crab_> zookeeper: or it's granted for the entire turn ?
20100428 22:13:16< gabba> Crab_: can you quote? I don't know what you're referring to by ^, I read it all :P
20100428 22:13:27< gabba> ahhh IRC...
20100428 22:13:29< Crab_> gabba: 'distract' ability
20100428 22:14:06< zookeeper> Crab_, check out The_Rise_Of_Wesnoth/utils/trow-abilities.cfg
20100428 22:14:12< gabba> distract? never heard of it
20100428 22:14:13< Crab_> zookeeper: thanks
20100428 22:14:17 * gabba goes to look
20100428 22:14:42< Soliton> just replay the actions to find conflicts that way you can't forget anything.
20100428 22:15:00< zookeeper> the ability engine takes care of it, the ability WML is basically just skirmisher with unconventional filtering and such to make it work on allies which are adjacent to the unit having the ability
20100428 22:15:24< Crab_> Soliton: yes, 'using the order of actions and pretending doing them' seems like a sane way to check
20100428 22:15:25< shadowmaster> pretty creative
20100428 22:16:22< zookeeper> yeah, and when i originally wrote that it IIRC exposed several bugs in the ability engine which needed to be fixed first :P
20100428 22:16:48< Crab_> gabba: what do you think about Soliton's idea ?
20100428 22:17:14< gabba> Crab_: "just replay the actions to find conflicts that way you can't forget anything" ?
20100428 22:17:51< Crab_> gabba: yes, basically check action #0 against current gamestate, #1 against the current+#0, #2 against current+#0+#1, etc
20100428 22:18:20< Crab_> by just copying the unit map and rewriting things on it, without any actual movement/attacks
20100428 22:19:19< Crab_> (so, there'll be no need to recalculate the entire srcdst map, just check the new route for units which actually move)
20100428 22:19:20< Soliton> also saving the whole path of an action would be nice if not too restricting. i think alink already implemented making a move with an explicit path and i for one would use it.
20100428 22:19:34< gabba> Crab_, Soliton: well first, I'm not sure if real "silent replays" exist yet: even fast replay in multiplayer animates the units
20100428 22:19:38< Crab_> Soliton: all moves are done with explicit path, internally
20100428 22:20:16< Crab_> gabba: "copy the unit map in C++, check if the move is valid, if yes, erase from old location and insert at new location." seems reasonably silent.
20100428 22:20:30< Soliton> Crab_: yeah, but it's important the interface allows specifying them completely or that is just annoying and leads to more conflicts.
20100428 22:21:27< gabba> Crab_: but if I fake moves, I have to basically rewrite some of the game code to fake the moves on my virtual map, so-to-speak
20100428 22:21:31< Crab_> Soliton: well, alink has disabled it because it was hard to control from UI side, yes ?
20100428 22:21:53< Soliton> possible, i don't know.
20100428 22:21:54< Crab_> gabba: you don't need to fake the moves, you can just move a unit on your fake map
20100428 22:22:19< Soliton> the ai already does that, no?
20100428 22:22:27< gabba> Crab_: ok, but then I don't see how I can easily take "distract" or any other random ability that might come up, into account
20100428 22:22:34< boucman> back...
20100428 22:22:37< boucman> hey gabba
20100428 22:22:46< gabba> hello boucman
20100428 22:22:49< Crab_> Soliton: no, but it has support for that, it's just not enabled.
20100428 22:23:09< Crab_> gabba: well, it will be internally and transparently taken care of by pathfinding code
20100428 22:23:26< gabba> boucman: big discussion ongoing with Crab_ and Soliton about conflict checking...
20100428 22:23:28< Crab_> gabba: you just need to use the pathfinding function which uses your fake unit map
20100428 22:24:20-!- Darkas [~quassel@ppp-93-104-156-119.dynamic.mnet-online.de] has joined #wesnoth-dev
20100428 22:24:26< Crab_> gabba: and pathfind::shortest_path_calculator, which can be passed to pathfind::a_star_search, already takes a unit map as a parameter.
20100428 22:24:57< Crab_> gabba: so, 'is move valid?' can be checked on any fake unit map, in a reasonably simple way
20100428 22:25:03< gabba> Crab_: so the pathfinder actually takes all units ability into account? nice
20100428 22:25:31< Crab_> gabba: yes
20100428 22:25:40< Crab_> grep the code for "skirmisher"
20100428 22:25:41-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100428 22:26:27< Crab_> gabba: the next piece is 'roughtly simulate the expected outcome of the move'. it's easy - either insert a traitless unit or insert a copy of unit from recall or move a unit on unit map.
20100428 22:28:16< Crab_> and the algorithm becomes :  foreach action { if action is valid, simulate it and continue, else mark conflicts and simulate it partially }
20100428 22:28:27< Crab_> where order in foreach is the order of actions
20100428 22:28:31< gabba> Crab_: ok, with your explanations "check action #0 against current gamestate, #1 against the current+#0, #2 against current+#0+#1" seems reasonable
20100428 22:28:46< Crab_> is_valid and simulate_it are already covered
20100428 22:29:27< Crab_> and 'simulate it partially' means "if a move has soft conflicts, then we mark both the src and dst as 'potentially occupied by that unit'"
20100428 22:29:48< Crab_> this has consequences of its own, of course
20100428 22:29:51-!- shadowm_bluecore [~ignacio@wesnoth/developer/shadowmaster] has quit [Remote host closed the connection]
20100428 22:30:11< Crab_> do we need to ensure that an earlier conflict doesn't mask latter conflicts ?
20100428 22:31:16< Crab_> those are second-order conflicts. "we want to move to A which is occupied, but the unit at A might move away (it has a planned move with soft conflicts on it)"
20100428 22:32:02< Crab_> anyway, that part can be skipped for simpilicity, as it can be handled later without disturbing the algorithm much.
20100428 22:32:10< timotei> mhh, why does wesnoth-git.tar contain inside another archive?
20100428 22:32:17< timotei> Isn't one step or archiving enough?
20100428 22:32:34< Crab_> timotei: tar.gz is a 'unite into one file and archive' 
20100428 22:32:45< Crab_> timotei: so, two operations
20100428 22:32:50< timotei> oh
20100428 22:33:02< boucman> gabba: from a code timeline PoV how possible would it be to implement without taking conflict into account at all, to simplify a first approch ?
20100428 22:33:30< Crab_> boucman: soft conflicts are just optional highlights
20100428 22:33:35< gabba> Crab_: I'm still hesitant though, because look at the extra complexity we need to add to cover basically distract and teleport!
20100428 22:33:59< Crab_> gabba: well, not just 'distract and teleport', but any WML ability that we can imagine or add later.
20100428 22:34:21< Crab_> gabba: 'reusing the pathfind code' means we'll cover everything that it can cover.
20100428 22:35:03< boucman> gabba: if you do conflict, i'm afraid using pathfind to detect conflict is the only sane way to go....
20100428 22:35:13< boucman> every other way lies madness :P
20100428 22:35:35< gabba> boucman: I'm getting ahead of myself here. We got into discussing it because Crab_ was looking at my class diagram (with conflicts) from my post on the ML
20100428 22:35:45< timotei> so based on what wikipedia says: tar is only for "holding" together some files, whilst gz/bz2 is used for compressing
20100428 22:35:51< timotei> interesting 
20100428 22:39:12< gabba> boucman, Crab_: well, I still think that just checking if the player's trying to send two units in the same hex would cover 90% of cases and would be very quick to implement, with no real possibility of bugs. What Crab_'s proposing is the smart and complete approach, but also sounds like something I should save for the end of the project
20100428 22:39:40< boucman> gabba: there are two types of mistakes you want to avoid with pathfinding
20100428 22:40:06< boucman> 1) offering a target position that can't be reached but won't be detected before execute
20100428 22:40:20< timotei> Crab_, I downloaded that git file, but no "svn files in there"
20100428 22:40:21< boucman> 2) not offering a position that would be possible
20100428 22:40:27< timotei> only the .git folder
20100428 22:40:42< Crab_> timotei: great
20100428 22:41:03< boucman> both 1 and 2 are highly dependant on ZoC, which is handled by the pathfinder...
20100428 22:41:09< gabba> boucman: what you mention sounds more related to hard conflicts
20100428 22:41:43< boucman> gabba: could you repeat the def of soft vs hard conflict for me plz ?
20100428 22:41:47-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100428 22:41:49< gabba> boucman: what I was discussing with Crab_ was about conflicts between the player's own moves (and his allies', when that comes into play)
20100428 22:42:11< gabba> boucman: ^that would be a soft conflict
20100428 22:42:35< Soliton> timotei: kind of expected since it's git...
20100428 22:42:47< gabba> boucman: a hard conflict is when the move is blocked by something outside the player's control, i.e. enemy units, cave wall, etc
20100428 22:43:01< Soliton> timotei: if you mean you miss the checkout then do "git reset --hard".
20100428 22:43:04< timotei> Soliton, yes. but I expected the files too
20100428 22:43:07< timotei> ok
20100428 22:43:18< boucman> so hard conflict is with ennemy/allied  units (assumed not to move during your turn) and soft conflict are with your own units (which depend on other planned orders) is that it ?
20100428 22:43:31< Crab_> boucman: almost.
20100428 22:43:53 * boucman had a bit too much alcohol to think clearly tonight
20100428 22:44:10< Crab_> boucman: there is one corner case: the first move to be executed is a hard conflict if it's not possible to do now, for whatever reason.
20100428 22:44:11< gabba> boucman: yes I forgot to mention allied unit that can block your movements, even though their ZOC doesn't affect you
20100428 22:44:46< gabba> Crab_: I haven't had any alcohol and I don't get it either :P
20100428 22:44:48< boucman> but allied units don't move during your turn, so they are hard conflicts, am I right ?
20100428 22:45:04< boucman> Crab_: I don't understand your corner case...
20100428 22:45:19< gabba> Crab_: yeah, the corner case... *cough*
20100428 22:46:01< Crab_> boucman: our actions are ordered, 1,2,3, etc
20100428 22:46:09< Crab_> boucman: that the order in which we'll do them
20100428 22:46:18< boucman> k
20100428 22:46:30< Crab_> boucman: actions which lead to units ending up in walls are always wrong, so they're 'hard'
20100428 22:46:49< Crab_> boucman: or in the enemy units
20100428 22:47:14< Crab_> boucman: also, the first action to be executed is hard conflict if we cannot do it now
20100428 22:47:34< Soliton> what's special about that?
20100428 22:47:51< boucman> Soliton: since it's the first action we are "sure" of the gamestate
20100428 22:47:55< Crab_> boucman: the rest of conflicts is soft 
20100428 22:48:11< boucman> Crab_: I get it now, not sure if it's worth special casing, but I understand it...
20100428 22:48:35< Crab_> boucman: it will be a special case anyway since it blocks the 'do next action' hotkey
20100428 22:48:47< gabba> Crab_: don't forget that action can be commited by the player in any order
20100428 22:49:49< timotei> Soliton, that checkout get the files from the "git staging right?
20100428 22:49:51< Crab_> gabba: yes, and that adds a group of second-order hard conflicts, as well.
20100428 22:50:12< gabba> I think hard conflicts should be individually checked per-action on an equal footing
20100428 22:50:13< Soliton> timotei: no idea what you just asked.
20100428 22:50:16< boucman> gabba: i'd like to have a working prototype as soon as possible, with whatever subset allows us to test the conflict and add some stuff easily
20100428 22:50:35< timotei> the git reset --hard
20100428 22:50:37< gabba> we assume that the gamestate can't change, and check every action against that
20100428 22:51:09< boucman> removing network is an important first step, removing soft conflict (which are hard to detect would be a nice step) keep in mind that the point is to get the concepts in place, and add them later
20100428 22:51:57< Crab_> timotei: yes
20100428 22:51:57< gabba> boucman: yes, getting something concrete up is a priority
20100428 22:52:09< timotei> weird, rebasing doesn't work
20100428 22:52:10< Soliton> timotei: that resets the checkout and since we don't include the checkout in the tarball it pretty much creates it.
20100428 22:52:27< timotei> I used: git svn rebase
20100428 22:52:28< Crab_> timotei: are you following a branch to rebase to ?
20100428 22:52:35< timotei> well, just trying to update
20100428 22:52:38< timotei> from the svn
20100428 22:52:45< Crab_> timotei: to what you want to update :) ?
20100428 22:52:53< timotei> oh, ye, you're right:D
20100428 22:53:03< gabba> boucman: can I go over your answer to the ML now?
20100428 22:53:14< boucman> gabba: especially since soft conflict won't have much influence on the overall ergonomy (just another type of arrow) so I would skip it for the moment
20100428 22:53:20< boucman> gabba: sure, good idea
20100428 22:53:44< Crab_> boucman: well, we started to talk about (soft) conflicts because we want to figure out a structure for storing the info
20100428 22:53:59< boucman> Crab_: ok,
20100428 22:54:00< Crab_> boucman: the detection algorithm is in the future, of course
20100428 22:54:14< boucman> ok, so we agree on the basic principles
20100428 22:54:23< Crab_> boucman: but we need to see if our current structure seems good enough for it to work
20100428 22:54:36< timotei> git svn fetch -rHEAD and I got some things like: 
20100428 22:54:54-!- Skywork [~croselius@pool-70-111-210-233.nwrk.east.verizon.net] has joined #wesnoth-dev
20100428 22:54:54< timotei> warning: LF will be replaced by CRLF in .git/MSRsHgZeMZ
20100428 22:54:54< timotei>         M       po/wesnoth-utbs/it.po
20100428 22:54:54< timotei> warning: LF will be replaced by CRLF in .git/MSRsHgZeMZ
20100428 22:54:54< timotei>         M       po/wesnoth-utbs/ja.po
20100428 22:54:55-!- Skywork is now known as Skystrikeer
20100428 22:54:58-!- Skystrikeer is now known as Skystriker
20100428 22:55:00< boucman> gabba: as I told earlier i won't be available form 1 evening to 9 morning, so I want to make sure you're on the right track before that
20100428 22:55:14< boucman> (which also explains the coding task in my mail
20100428 22:55:27< Crab_> as we see from the Soliton's idea, we'll "collect" those conflicts 'after' going through the actions one by one
20100428 22:55:39< Crab_> so, we can indeed store a list of conflicts in each action
20100428 22:56:11< gabba> Crab_, boucman: I'll start with storing hard conflict status directly in planned actions (it's a binary state), and move them out in a more general conflicts structure if needed when I get to soft conflicts. Sounds ok?
20100428 22:56:40< boucman> gabba: stupid question, why do you even allow to plan hard conflict ?
20100428 22:56:46< Soliton> timotei: no idea about that. probably some windows issue. maybe you can just ignore it.
20100428 22:56:50< timotei> ok
20100428 22:57:11< gabba> boucman: plan: never, but wml events could create some by spawning walls or what not
20100428 22:57:23< Crab_> gabba: yes. be sure to reuse the ai code for checking hard conflicts.
20100428 22:57:29< gabba> Crab_: yes
20100428 22:57:46< timotei> oh, so enjoying... no more .svn directories in each directory
20100428 22:57:54< Crab_> boucman: 'move into the shroud', or move into a hex that will contain enemy ambusher.
20100428 22:58:04< boucman> gabba: not sure what you mean, if WML event changes the layout, you can't take it into account, so you would still not offer hard conflicts to the players, 
20100428 22:58:07< Crab_> boucman: or wml terrain changes
20100428 22:58:32< Crab_> boucman: you plan 5 moves, execute 1, terrain changes or units are spotted, hard conflicts appear
20100428 22:58:44< Soliton> timotei: does "git status" mention those files?
20100428 22:58:53< boucman> Crab_: and the remaining 4 moves become hard conflict...
20100428 22:58:57< boucman> ok, I get it...
20100428 22:59:00< Crab_> boucman: yes
20100428 22:59:16< Crab_> boucman: (assuming each of those 4 moves is not valid anymore)
20100428 22:59:20< gabba> boucman: what Crab_ said, I was typing the same thing :)
20100428 22:59:22< boucman> in that case I would simply drop the conflicting moves entirely, but I can understand everybody wouldn't agree to that...
20100428 22:59:31< timotei> Soliton, did a simple diff on the directory: and have plently of: diff --git a/SConstruct b/SConstruct
20100428 22:59:31< timotei> old mode 100755
20100428 22:59:31< timotei> new mode 100644
20100428 22:59:36< Crab_> boucman: no, we shouldn't drop them
20100428 22:59:38< timotei> on different files:|
20100428 23:00:15< Soliton> timotei: google or talk to ilor about such windows issues.
20100428 23:00:31< timotei> ok:)
20100428 23:00:35< timotei> thanks anyway for help
20100428 23:00:43< boucman> Crab_: yes and no, they don't make sense anyway...
20100428 23:00:43< timotei> I'll go now
20100428 23:00:48< timotei> good night everyone
20100428 23:00:54< gabba> boucman: dropping them or not isn't what complicates the code anyways
20100428 23:00:58< gabba> night timotei
20100428 23:01:02< boucman> true
20100428 23:01:12< Lastmerlin> question: does anyone know a FOSS, highlevel network abstraction library ?
20100428 23:01:55< Crab_> boucman: why ? if a move becomes impossible due to an enemy, the player might still want to move to nearby hex. after all, we shouldn't ruin planning - we should allow the player to decide what he'll do and how he'll change
20100428 23:02:11< Lastmerlin> btw, what does wesnoth use here ?
20100428 23:02:27< boucman> Crab_: simplification, at least as a first draft...
20100428 23:02:36< Crab_> Lastmerlin: sdl.net and switching to custom library which will use boost::asio
20100428 23:02:52< Crab_> boucman: I agree about 'not coding everything from the start', yes
20100428 23:02:54< boucman> but yes on principle, and if it doesn't conflict the UI (i.e not too much info when compared to soft conflicts) why not
20100428 23:03:09< gabba> boucman: ok, if Crab_ doesn't mind ( ;) ) I'll go over your ML post now.
20100428 23:03:18< Crab_> gabba: of course :)
20100428 23:03:24 * Crab_ disappears
20100428 23:03:26< Lastmerlin> so,  selfmade library based on boost, essentially ?
20100428 23:03:30< boucman> i'm a bit afraid we would have too much type of arrows or non-sensical arrows (planned moves going through wall would be hard to make sense of)
20100428 23:03:35< Crab_> Lastmerlin: yes
20100428 23:03:57< Crab_> Lastmerlin: plans are to make it generic enough to be reused.
20100428 23:04:29< Lastmerlin> hmm it seems, that I have to change network lib for my project now x(
20100428 23:05:13< gabba> boucman:  "to have a getChanges, you need to somehow know "since when"": agreed, I was thinking of keeping track in PlannedActionSet of what was sent yet
20100428 23:05:35< Crab_> Lastmerlin: why :) ?
20100428 23:05:39 * boucman goes back to read his mail :)
20100428 23:05:43-!- Sirp__ [~97c1dc1b@wesnoth/developer/dave] has quit [Quit: Page closed]
20100428 23:05:57< gabba> boucman: anyways just knowing where the function will live is enough for now, since networking is optional
20100428 23:05:59< Lastmerlin> it changed license and does not offer GPL anymore
20100428 23:06:05< Lastmerlin> and I did not notice
20100428 23:06:11< Crab_> Lastmerlin: ah
20100428 23:06:14< boucman> gabba: ok, make sense...
20100428 23:06:14< Lastmerlin> nor did I save a GPLed version
20100428 23:06:22< Lastmerlin> which is a real headache x(
20100428 23:06:32< gabba> boucman: here's the uml diagram link for easy reference btw: http://forums.wesnoth.org/download/file.php?id=40412&mode=view
20100428 23:06:50< boucman> i'm not sure keeping the info in the structure makes sense (mutiple desynchronized targets) but at this point it doesn' tmatter
20100428 23:06:55< boucman> thx
20100428 23:07:13< Lastmerlin> hrm sdl net is an ugly set of functions, that barely wrap sockets
20100428 23:07:43< gabba> boucman: about exposing or not the stl vector, I still hesitate on this point
20100428 23:08:07< gabba> boucman: the idea was to be able to change easily to a deque (for instance) if needed
20100428 23:08:34< Lastmerlin> what I had so far, was a neat, object oriented lib with nice interface
20100428 23:08:53< Lastmerlin> that is what I meant by highlevel
20100428 23:09:10< gabba> boucman: I can reimplement an iterator and everything, but I expect a few bugs before I get all the operations right
20100428 23:09:20-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has joined #wesnoth-dev
20100428 23:09:25< norbert_> Sirp, are you there?
20100428 23:09:35-!- Blueblaze [~nick@adsl-76-202-23-150.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 260 seconds]
20100428 23:09:35< boucman> gabba: hmm, changing from vector to deque if the vector is exposed is ~1/2 day work at most, and the bugs could get annoying... i'd go for exposing the vector
20100428 23:09:52< boucman> moreover I can't really see why you would want a deque... 
20100428 23:10:29< gabba> deque is fast for insertions at the end, and that's what we're gonna do all the time with planned actions
20100428 23:10:50< Crab_> gabba: the time for insertion at the end will be o(1), mostly
20100428 23:10:53< boucman> true
20100428 23:10:56< gabba> insertions in a vector are slow compared deque and list
20100428 23:11:07< gabba> s/compared/compared to
20100428 23:11:11< Crab_> gabba: the number of planned actions is not big.
20100428 23:11:19< boucman> ok, then why would you want vectors, and why not expose the deque ? :P
20100428 23:11:43< Crab_> gabba: for example, preallocate size=100 in the vector and in 99.99% cases you won't hit that limit.
20100428 23:12:13< gabba> boucman: yes, I could go for a deque right away. but then it's apparently less iterator-safe than vectors, so I wanted to avoid having to do the heavy research right now
20100428 23:12:48< gabba> Crab_: so, you'd say go for vector or deque?
20100428 23:13:05< Crab_> gabba: basically, your operations to work on that list of planned actions will be very cheap compared to things like A* in pathfinding for validity checking
20100428 23:13:17-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 252 seconds]
20100428 23:13:18< Crab_> gabba: I'd say "go for simplest thing that will work"
20100428 23:14:10< gabba> Crab_: yeah, seen in light of your comparison...
20100428 23:14:18< boucman> vector have o(1) indexed access, which dequeues don't have... that's a good reason to use vectors
20100428 23:14:38< Crab_> boucman: deque has o(1) indexed access, afair...
20100428 23:14:57-!- timotei [timotei@188.24.7.149] has quit [Ping timeout: 258 seconds]
20100428 23:15:06-!- Zarel [~Zarel@warzone2100/developer/Zarel] has quit [Ping timeout: 245 seconds]
20100428 23:15:38< gabba> http://www.cppreference.com/wiki/stl/deque/start
20100428 23:15:40< boucman> Crab_: true, my bad...
20100428 23:16:52-!- Lastmerlin [~Lastmerli@kalypso.csn.tu-chemnitz.de] has left #wesnoth-dev ["Kopete 0.12.7 : http://kopete.kde.org"]
20100428 23:17:21-!- Lastmerlin [~Lastmerli@kalypso.csn.tu-chemnitz.de] has joined #wesnoth-dev
20100428 23:17:53-!- norbert_ [~norbert@82-171-70-54.ip.telfort.nl] has quit [Quit: Leaving]
20100428 23:19:08< boucman> gabba: any question about the coding task ?
20100428 23:19:26< gabba> deque lacks capacity() and reserve(), which I won't use, so I think I should use that... it doesn't seem to have any glaring disadvantage. So that's settled for now.
20100428 23:19:32< gabba> boucman: getting there
20100428 23:19:39< boucman> k
20100428 23:21:15< gabba> boucman: the separate draw functions were because I envision having them on different layers, I was under the impression (maybe false) that might imply adding them at different times.
20100428 23:21:36< boucman> hmm
20100428 23:21:47-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Ping timeout: 248 seconds]
20100428 23:22:29< gabba> maybe it's just "add everything to the right layer, let the drawing engine sort this out", it's too hazy for me as of now
20100428 23:22:44< boucman> well, for ghosts, I'd advise doing my coding task and then "registering" ghost units, rather than drawing them yourself... it's much simpler to reuse the unit drawing code and I have a good mastery of that area
20100428 23:22:55< boucman> for arrows I guess they would all be on the same layer
20100428 23:23:22< boucman> not sure if there is anything left (action number maybe ?)
20100428 23:24:07< gabba> boucman: action number and attack markers I guess
20100428 23:24:09< gabba> boucman: those fake units are very interesting. Do they show unit stats when you hover them
20100428 23:24:15< gabba> ?
20100428 23:24:37< boucman> they can, but usually don't 
20100428 23:24:49< boucman> hmm
20100428 23:25:00< boucman> well, xp/hp is optional, 
20100428 23:25:19< boucman> I don't think they accept hvering over, but that could be probably fixed
20100428 23:25:47< boucman> problem is what to display if there are more than one unit in a given hex (ghost+real)
20100428 23:26:57< gabba> boucman: yes, and also what to display for ghost recruits, since they don't have random traits yet, but the player might want to check them out (and maybe use the Wesnothpedia to get the description)
20100428 23:27:25< boucman> i don't know if wesnothopedia works on ghost
20100428 23:27:47< boucman> current ghost are meant to be "transiant" if you see what I mean, so that sort of issues havn't been sovled
20100428 23:28:12< boucman> i'd say that generate without trait (i.e base unit) since it's the only thing that makes sense...
20100428 23:28:39< Crab_> gabba: without traits, I'd say
20100428 23:28:41< gabba> boucman: I can add extra functionality to those fake units as part of the small coding task
20100428 23:28:52< Crab_> gabba: but, for recalls, you can set the correct traits
20100428 23:28:53< boucman> yes, it could be nice...
20100428 23:29:07< gabba> ok
20100428 23:29:38< gabba> boucman: the last thing (and I've gotta go):
20100428 23:30:13< gabba> I took note of the possibilities for getting start positions and all from units, also Crab_ and others mentioned some stuff about that
20100428 23:30:36< gabba> I'll adjust those as I go
20100428 23:30:59-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: Leaving.]
20100428 23:33:03-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100428 23:34:21< gabba> boucman: so for the next step...hmmm... I'm tempted to leave this data structure as an UML diagram until I actually need it (and then I click generate and fill up the blanks)
20100428 23:34:53< boucman> gabba: ok, makes sense...
20100428 23:35:03< boucman> so, plan for next week 
20100428 23:35:20< boucman> 1) do the multiple fake unit to get your bearing in the drawing code, 
20100428 23:35:50< boucman> 2) probably implement the data structure and if possible the related drawing code (hard coding a couple of moves to see what it displays
20100428 23:35:58< boucman> at which point i'll probably be back
20100428 23:36:03< boucman> something like that ?
20100428 23:36:16< gabba> boucman: Don't you think that this would be better:
20100428 23:36:34< gabba> 1) Get some basic arrow drawing (i.e. dots in the right hexes) up
20100428 23:36:59< gabba> 2) Link this with the data structure, so I can get the drawing framework up
20100428 23:38:33< gabba> 3) multiple fake units
20100428 23:38:43< boucman> gabba: if you prefer it in that order...
20100428 23:39:19< boucman> I think multiple fake is a good way to learn since it's basicaly replace a single object with a vector, then fix where it doesn't compile
20100428 23:39:42< gabba> boucman: I don't really mind, actually
20100428 23:39:55< boucman> but if you prefer doing it the other way round, that's fine too, 3 is more or less independant from 1+2 (codewise)
20100428 23:40:16< gabba> multiple fake units is a nice sub-task with a concrete start and end point
20100428 23:40:30< boucman> and possibility to merge immediately
20100428 23:40:50< Crab_> gabba: for arrows, be sure to see if there's a way to draw an arrow from any part of wesnoth's code, independently of planned actions context
20100428 23:41:11< gabba> Crab_: yes, framework approach and all that ;)
20100428 23:41:13< Crab_> gabba: e.g., having an 'draw an arrow from X to Y using style Z' will be a cool thing in many other places
20100428 23:41:31< Crab_> gabba: for example, I'd be able to trick the AI into using it for debugging purposes...
20100428 23:42:02< gabba> Crab_: that'd be nice
20100428 23:42:04-!- necropresto [~necropres@unaffiliated/necropresto] has joined #wesnoth-dev
20100428 23:42:05< boucman> so "registering an arrow" sort of thing ?
20100428 23:42:27< Crab_> boucman: don't know, depends on what is easier to do
20100428 23:42:35< gabba> well, I have to go
20100428 23:42:55< gabba> I have a Shadowrun game tonight, Crab_ you gotta know that
20100428 23:43:01< boucman> gabba: ok, will you be here on sunday ?
20100428 23:43:22< Crab_> gabba: :)
20100428 23:43:31< gabba> boucman: this w-e I won't be around much (not guaranteed)
20100428 23:43:50< boucman> k, then we will probably communicate mainly through email until I leave...
20100428 23:44:09< gabba> boucman: alright
20100428 23:44:18< boucman> please post a plan for next week on ML, and i'll reply tomorow... though our current plan seems fine overall
20100428 23:44:43< gabba> ok
20100428 23:44:57< gabba> see you all later :)
20100428 23:45:24-!- gabba [~gabriel@70.35.167.54] has left #wesnoth-dev []
20100428 23:46:19-!- necropresto [~necropres@unaffiliated/necropresto] has left #wesnoth-dev []
20100428 23:46:34-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100428 23:58:50< crimson_penguin> Ivanovic: looks like my fix worked
--- Log closed Thu Apr 29 00:00:48 2010