--- Log opened Mon Jul 27 00:00:24 2009
20090727 00:01:10-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090727 00:02:40-!- melinath [n=melinath@stud244178.studentenheim.uni-tuebingen.de] has joined #wesnoth-dev
20090727 00:09:14-!- wesbot changed the topic of #wesnoth-dev to: 66 bugs, 239 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090727 00:09:48-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has joined #wesnoth-dev
20090727 00:13:16< CIA-62> euschn * r37250 /trunk/src/playcampaign.cpp: removed obsolete assertion
20090727 00:21:44< boucman> night all
20090727 00:21:54-!- boucman [n=rosen@wesnoth/developer/boucman] has quit ["Leaving."]
20090727 00:36:26-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz"]
20090727 00:36:51-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Remote closed the connection]
20090727 00:40:31-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090727 00:40:57-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090727 00:42:32-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has quit ["And that’s the end of THAT chapter."]
20090727 00:48:19-!- AFKMJ [n=msflower@80.154.28.81] has quit [Read error: 104 (Connection reset by peer)]
20090727 00:48:29< crimson_penguin> well, the 1.6 server going down sure got people playing 1.7 :D
20090727 00:49:32-!- ancestral [n=ancestra@166.205.6.24] has joined #wesnoth-dev
20090727 00:51:34< AI0867> ilor: there is one issue with splitting up recruit_unit() into two parts: the first thing it does is create a command_disabler. Moving it into each function creates a race condition while forcing the caller to take care of it is a whole problem of its own
20090727 00:51:53-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has quit [Remote closed the connection]
20090727 00:52:15< Soliton> crimson_penguin: all versions went down. ;-)
20090727 00:52:55< ilor> AI0867: forcing the caller to do it seems like the only choice. Not like many places use the code anyway (I'm guessing 2)
20090727 00:54:06< ilor> AI0867: figuring out a robust way of doing it is a pain esp. if we want to introduce a lenghty wait for a server response in the middle. The way it's done for combats now is far from bullet-proof, sadly.
20090727 00:54:12-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090727 00:54:42< crimson_penguin> Soliton: well then there was a bunch of people (including me), who couldn't connect to 1.6, and when they tried to connect to 1.7, it did work
20090727 00:57:35< Soliton> crimson_penguin: not saying you're wrong. just a bit of a misinformed dynamic then. :-)
20090727 00:57:44< crimson_penguin> yeah :)
20090727 00:58:03< crimson_penguin> but when I got there, there was like 1 person, and then suddenly 2 more people joined, and later there were more
20090727 00:59:18-!- MJ999 [n=msflower@80.154.28.81] has joined #wesnoth-dev
20090727 01:00:02< Soliton> crimson_penguin: not really more than before though if you check here: http://wesnothd.wesnoth.org
20090727 01:00:26< crimson_penguin> yeah, it wasn't a huge influx, but I'm pretty sure that was the reason
20090727 01:07:39-!- ancestral [n=ancestra@166.205.6.24] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090727 01:07:42-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090727 01:08:04-!- ancestral [n=ancestra@166.205.6.24] has joined #wesnoth-dev
20090727 01:18:49< CIA-62> ai0867 * r37251 /trunk/data/multiplayer/scenarios/2p_Dark_Forecast.cfg: Ran wmlindent on dark forecast.
20090727 01:20:11-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has joined #wesnoth-dev
20090727 01:33:48-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit ["Quitte"]
20090727 01:38:35< CIA-62> ai0867 * r37252 /trunk/data/multiplayer/scenarios/4p_A_New_Land.cfg: Fix a typo in a wmllint pragma.
20090727 01:39:25-!- Chusslove [n=Chusslov@brsg-d9befabf.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20090727 01:44:58-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090727 01:46:30-!- Chusslove [n=Chusslov@brsg-d9befe82.pool.mediaWays.net] has joined #wesnoth-dev
20090727 01:47:09-!- elias [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090727 01:49:04-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090727 01:54:52-!- ancestral [n=ancestra@166.205.6.24] has quit ["Get Colloquy for iPhone! http://mobile.colloquy.info"]
20090727 02:02:04-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection reset by peer]
20090727 02:02:19-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 02:02:44-!- mode/#wesnoth-dev [-bb *!*@irseek/log-bot* zenzei2!*@*] by ChanServ
20090727 02:08:55-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has joined #wesnoth-dev
20090727 02:09:57< CIA-62> ai0867 * r37253 /trunk/data/tools/wmlindent: Teach wmlindent about <<lua strings>>.
20090727 02:28:09-!- ardesh_ [n=ardesh@port-92-206-75-230.dynamic.qsc.de] has joined #wesnoth-dev
20090727 02:30:22-!- ardesh [n=ardesh@port-92-206-40-230.dynamic.qsc.de] has quit [Read error: 113 (No route to host)]
20090727 02:31:38-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090727 02:31:52-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090727 02:43:03-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090727 02:43:53-!- Appleman1234 [n=Appleman@131.181.102.53] has quit [Read error: 110 (Connection timed out)]
20090727 02:44:41-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has left #wesnoth-dev []
20090727 02:45:03-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090727 02:48:29-!- Espreon [n=espreon@wesnoth/developer/espreon] has quit ["WRYYYYYYYYYYYYYYYYYYYY!"]
20090727 02:49:51-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090727 02:52:05-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 02:58:21-!- Aethaeryn is now known as Aythayryn
20090727 02:58:25-!- Zarel is now known as Zaryl
20090727 02:59:32-!- ancestral is now known as ancestryl
20090727 02:59:51-!- melinath [n=melinath@stud244178.studentenheim.uni-tuebingen.de] has quit [Read error: 110 (Connection timed out)]
20090727 02:59:53-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090727 03:03:57-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has left #wesnoth-dev []
20090727 03:21:19-!- Aythayryn is now known as Aethaeryn
20090727 03:28:19-!- Appleman1234 [n=Appleman@131.181.102.53] has joined #wesnoth-dev
20090727 03:32:09-!- ancestryl is now known as ancestral
20090727 04:03:27-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has quit []
20090727 04:07:10-!- Ivanovic_ [n=ivanovic@dtmd-4db2c7ad.pool.einsundeins.de] has joined #wesnoth-dev
20090727 04:08:32-!- Kenpachi [n=chatzill@121.220.195.178] has joined #wesnoth-dev
20090727 04:14:40-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has joined #wesnoth-dev
20090727 04:24:08-!- Ivanovic [n=ivanovic@wesnoth/developer/ivanovic] has quit [Read error: 110 (Connection timed out)]
20090727 04:25:08-!- Ivanovic_ is now known as Ivanovic
20090727 04:28:49< ancestral> crimson_penguin: I'm getting a CTD on 1.7.2 when I change the resolution to certain sizes
20090727 04:28:55< CIA-62> ai0867 * r37254 /trunk/src/ (playsingle_controller.cpp replay_controller.cpp): Remove unneeded try-catch.
20090727 04:29:01< CIA-62> ai0867 * r37255 /trunk/src/ (playsingle_controller.cpp playsingle_controller.hpp): Make [end_turn] work in new turn, side turn and turn refresh [event]s again.
20090727 04:29:16< ancestral> When not full screen. I understand there might be some full screen issues with SDL, but it would seem that maybe this is unrelated
20090727 04:29:47< ancestral> When not in full screen, when I change resolution to a size larger than my screen size, it crashes
20090727 04:30:09< Turuk> really?
20090727 04:30:17< Turuk> Huh, I was playing with that today and it worked for me
20090727 04:30:21< Turuk> what are you switching to?
20090727 04:30:44< ancestral> Oddly, when I choose 1344x840 it crashes
20090727 04:30:51< ancestral> which is less than 1440x900
20090727 04:31:09< ancestral> 1.6 and 1.7 both
20090727 04:31:24 * Turuk shrugs
20090727 04:31:30< ancestral> Working for you?
20090727 04:31:32< Turuk> I don't have that as an option to switch to
20090727 04:31:53< ancestral> What if you try switching to the screen resolution when not in full screen?
20090727 04:32:22< Turuk> works just fine
20090727 04:32:41< Turuk> where'd you get 1344x840 from?
20090727 04:35:03< ancestral> It's listed in 1.7.2
20090727 04:35:17< ancestral> Actually in 1.6 also
20090727 04:35:24< ancestral> Might be the version of MBP I ave
20090727 04:35:32< ancestral> Yeah looks like it
20090727 04:37:24< ancestral> Still is odd…
20090727 04:38:00< ancestral> When I switch to 1280x800 I get a partially rendered window
20090727 04:39:49< Turuk> I don't get those issues
20090727 04:40:00< ancestral> Interesting
20090727 04:40:57< ancestral> Where's the sdl bug tracker?
20090727 04:41:26< ancestral> Ahhh found it
20090727 04:44:40< ancestral> Interesting
20090727 04:45:28< ancestral> Basically the guy who does most of the Mac work on SDL (it appears) is icculus
20090727 04:46:36< ancestral> He's done only a few things: http://icculus.org/
20090727 04:56:06< CIA-62> ai0867 * r37256 /trunk/src/playmp_controller.cpp: Make the playsingle changes in playmp too.
20090727 04:56:10< CIA-62> ai0867 * r37257 /trunk/src/ (4 files): Clean up some unneeded try-catch statements.
20090727 04:57:23-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Success]
20090727 04:57:38-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 05:15:12-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 60 (Operation timed out)]
20090727 05:15:50-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 05:25:06< Aethaeryn> If the faction is fixed by WML, but not the leader, will someone who joins as say, loyalists and lieutenant, not be able to keep the choice of lieutenant?
20090727 05:26:06< Aethaeryn> Hmm, just tested it out, and it gives you random leader. Perfect.
20090727 05:47:21-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090727 05:47:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 05:49:01-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"]
20090727 05:55:11-!- Chusslove [n=Chusslov@brsg-d9befe82.pool.mediaWays.net] has quit [Read error: 110 (Connection timed out)]
20090727 06:00:12-!- Chusslove [n=Chusslov@brsg-d9bee139.pool.mediaWays.net] has joined #wesnoth-dev
20090727 06:08:09< crimson_penguin> ancestral: CTD? oh, right, crash to desktop
20090727 06:08:35< ancestral> It's Mac specific it sounds, and it sounds like only on some machines?
20090727 06:08:40< ancestral> Probably is SDL
20090727 06:09:12< ancestral> Reproducible every time for me. Might have something to do with the video card in the new MBP's
20090727 06:09:14-!- wesbot changed the topic of #wesnoth-dev to: 67 bugs, 239 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090727 06:09:16< ancestral> not sure
20090727 06:09:50< crimson_penguin> I actually reported a similar bug before... I forget what it was though
20090727 06:09:58< crimson_penguin> But it only happened at specific sizes
20090727 06:15:04< ancestral> Yes
20090727 06:15:19< ancestral> 1344x840 and 1440x900
20090727 06:15:37< ancestral> CTD every time, when not full screen… though actually I haven't tried it with full screen now
20090727 06:25:54-!- crimson_penguin [n=ben@wesnoth/developer/crimsonpenguin] has quit []
20090727 07:00:14-!- Appleman1234 [n=Appleman@131.181.102.53] has quit [Read error: 104 (Connection reset by peer)]
20090727 07:14:08-!- silene [n=plouf@ASte-Genev-Bois-152-1-54-203.w82-121.abo.wanadoo.fr] has joined #wesnoth-dev
20090727 07:23:50-!- Netsplit brown.freenode.net <-> irc.freenode.net quits: Mellar_, ettin, corn, Kenpachi, AI0867
20090727 07:23:52-!- Sirp [n=user@wesnoth/developer/dave] has quit ["leaving"]
20090727 07:28:23-!- Netsplit over, joins: corn, AI0867, Kenpachi
20090727 07:28:47-!- Netsplit over, joins: ettin, Mellar_
20090727 07:36:39-!- Kenpachi [n=chatzill@121.220.195.178] has quit [Read error: 110 (Connection timed out)]
20090727 07:42:12-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has quit []
20090727 07:48:36< Blueblaze> Does anyone know how I could subtract an x coordinate within the x= line? For example x=$x1+1 would work fine, but x=$x1-1 wouldn't work
20090727 07:49:04< shadowmaster> x="$($x1 + 1)"
20090727 07:49:14< shadowmaster> er, -
20090727 07:49:34< shadowmaster> your first example doesn't do what you think it does
20090727 07:50:17< Blueblaze> What did my first example do? Since it did work properly
20090727 07:50:56< shadowmaster> it should get converted into x=$x11 unless you were using Lua or something else that's not regular WML
20090727 07:58:05< Blueblaze> shadowmaster: Any reason why x="$($x1 - 1)" places the action to the right of the $x1 (thus increasing it by 1) and x="$($x1 + 1)" places the action to the left of the $x1 (thus decreasing it by 1)
20090727 07:58:11< Blueblaze> Seems backwards to me, but it does work
20090727 07:59:15< Blueblaze> Oh, nevermind, I understand why
20090727 07:59:35< Blueblaze> I'm using it for a [show_if] relative to the position of a unit, so of course it will be "backwards"
20090727 08:03:34-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has quit ["'night"]
20090727 08:12:14-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 08:19:29-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Connection timed out]
20090727 08:20:23-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 08:23:06-!- zookeeper [n=l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20090727 08:23:47-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Connection reset by peer]
20090727 08:24:17-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 08:24:33-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 08:32:28-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 08:34:39-!- Kenpachi [n=chatzill@CPE-121-220-195-178.sa.bigpond.net.au] has joined #wesnoth-dev
20090727 08:38:19-!- melinath [n=melinath@vpn1066.extern.uni-tuebingen.de] has joined #wesnoth-dev
20090727 08:40:02-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 08:46:35-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 08:47:30-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 08:56:54-!- euschn [n=chatzill@wesnoth/developer/euschn] has joined #wesnoth-dev
20090727 08:56:58< euschn> hi
20090727 08:58:08< euschn> Ivanovic: I read on yesterdays logs you had some assertion failure after playing the first scenario of AToB - was that in trunk or in a 1.7.x tag?
20090727 09:03:18-!- Elvenisko [n=JA@ig511.imafex.sk] has joined #wesnoth-dev
20090727 09:04:03< Elvenisko> hello everyone, i havea great problem
20090727 09:04:17< Elvenisko> i made a campaign in the past, called the three elves
20090727 09:04:29< Elvenisko> can i found that data on the internet?
20090727 09:04:34< Elvenisko> it was for wesnoth 1.4
20090727 09:05:15-!- stikonas [n=and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20090727 09:05:23< Blueblaze> add-ons.wesnoth.orrg
20090727 09:05:26< Blueblaze> add-ons.wesnoth.org
20090727 09:05:33< Turuk_> www.wesnoth.org/addons/1.4
20090727 09:06:07< Blueblaze> http://files.wesnoth.org/addons/1.4/The_Three_Elves.tar.bz2
20090727 09:06:10< Blueblaze> :P
20090727 09:06:20< Elvenisko> great many many many thanks :)
20090727 09:06:54< Turuk_> www.wesnoth.org/Never/Listen/To/Blueblaze
20090727 09:07:20< Blueblaze> Not Found, The requested URL /Never/Listen/To/Blueblaze was not found on this server. :(
20090727 09:07:35< Blueblaze> Turuk_: Work on fixing that asap
20090727 09:08:16< Elvenisko> maybe server is offline :)
20090727 09:15:38-!- silene [n=plouf@wesnoth/developer/silene] has quit ["Leaving."]
20090727 09:23:28-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 104 (Connection reset by peer)]
20090727 09:25:34-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090727 09:26:48-!- melinath [n=melinath@vpn1066.extern.uni-tuebingen.de] has quit [Read error: 104 (Connection reset by peer)]
20090727 09:26:53-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090727 09:27:03< Crab_> hi
20090727 09:27:34-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Remote closed the connection]
20090727 09:27:48-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 09:27:54-!- loonybot [n=loonybot@79.139.136.93] has joined #wesnoth-dev
20090727 09:34:03< euschn> hi Crab_ 
20090727 09:34:36< Crab_> hi euschn
20090727 09:34:39< Crab_> Dragonking: around ?
20090727 09:36:35-!- Blueblaze [n=nick@c-98-199-143-139.hsd1.tx.comcast.net] has quit [Remote closed the connection]
20090727 09:39:09< euschn> ilor: you made some changes to rng recently, correct?
20090727 09:41:16-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090727 09:55:16-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has joined #wesnoth-dev
20090727 09:55:23-!- ancestral [n=ancestra@97-116-106-168.mpls.qwest.net] has quit [Remote closed the connection]
20090727 09:57:41< Dragonking> Crab_: I am now
20090727 10:00:39< Crab_> Dragonking: hello. I've begin to do this 'split fai initialization into two parts' thing, and there's some caveats there. Basically, I initialize fai at the point where only readonly_context is available, but fai inherits from ai_default which needs default_ai_context to construct. 
20090727 10:00:50< ilor> euschn yeah
20090727 10:00:58< Crab_> the solution I've found is to create a special proxy of default_ai_context which will throw an initialization_exception if any unavailable function is called, and initialize fai using it
20090727 10:01:27< Dragonking> Yes, makes sense
20090727 10:01:50< Crab_> (and then, later, during the second part of initialization, switch the proxy's target to a 'real' default_ai_context
20090727 10:02:06< Dragonking> ok:)
20090727 10:05:03< Dragonking> Crab_: I've been working on this 'map possession calculation'
20090727 10:05:42< Crab_> and how it is ?
20090727 10:05:52< Dragonking> Crab_: I tried 'brute force' approach, iterating over units, using BFS and somply movement cost as 'score'
20090727 10:06:01< Dragonking> It is too slow
20090727 10:06:21< Dragonking> On my dual core machine it takes 8s for 70x70map with 100 units
20090727 10:06:46< Crab_> have you profiled ?
20090727 10:06:53< Dragonking> Not really
20090727 10:07:11< Crab_> it would be interesting to see 'why it needs this time'.
20090727 10:07:46< Crab_> for there are cases where some cache or content-addressable lookup table would help
20090727 10:08:18< Dragonking> But cache of what?
20090727 10:08:59< ilor> euschn: best just write what the problem is, people generally read logs here :)
20090727 10:09:07< Dragonking> Crab_: I used modified find_routes from pathfind.cpp
20090727 10:10:42< Crab_> what is the modification ?
20090727 10:11:32< Crab_> I was thinking about something like 'once for each unit type, fire floyd-warshall to collect all possible distances, then cache the results'
20090727 10:11:38< Dragonking> I removed some parameters - like assumed always ignore_units=true and thus removed some code, plus it runs on the whole map, not only on a part like find_routes was
20090727 10:12:16< Dragonking> Crab_: Yes, tha't something that came to my mind too
20090727 10:12:41< Dragonking> Crab_: Not exectly floyw-warshall but something around this
20090727 10:12:47< Crab_> note that it can be cached in some kind of global cache
20090727 10:14:04< Crab_> for example, human mp player knows (roughly), what places are reachable 'first' by each player on standard maps
20090727 10:14:30< Crab_> and if unit types / movement costs / maps do not chance, this info can be cached even between games.
20090727 10:14:44< Crab_> s/chance/change
20090727 10:15:10< Crab_> it can be even precalculated on wesnoth release.
20090727 10:15:18< Dragonking> Crab_: Well, for now I'm thinking only about random maps and some optimal solution
20090727 10:15:36-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 10:15:52< Crab_> Dragonking: well, how big % of games is done on random maps ?
20090727 10:16:13< Dragonking> Crab_: But the you owuld need to cache it for every map?
20090727 10:16:32< Crab_> Dragonking: yes
20090727 10:16:38< Dragonking> For every map, vache every unit id
20090727 10:16:46< Crab_> Dragonking: 'unit type'
20090727 10:17:28< Dragonking> Yes
20090727 10:17:32< Crab_> Dragonking: how big it will be this way ?
20090727 10:17:50< Dragonking> Taking onto account campaigns?
20090727 10:17:59< Dragonking> We have load of units
20090727 10:18:12-!- EdB [n=edb@171.117.88-79.rev.gaoland.net] has joined #wesnoth-dev
20090727 10:18:32< Ivanovic> euschn: trunk
20090727 10:18:44< Ivanovic> though your commit last night might have fixed it
20090727 10:18:53< Ivanovic> moin
20090727 10:19:13< Dragonking> Crab_: Plus still we do need way to calculate if fast for custom games/campaigns/scenarios
20090727 10:19:26< Dragonking> So I don't think that it is really worth to just cache everything
20090727 10:19:33< Crab_> Dragonking: fgrep for [unit_type] gives us (in mainline), 379
20090727 10:20:24< Dragonking> Including units from scenarios?
20090727 10:20:25-!- Zaryl [i=Zarel@warzone2100/developer/Zarel] has quit [Read error: 110 (Connection timed out)]
20090727 10:20:34< Crab_> yes
20090727 10:21:21< Crab_> let it be 400, then, 70*70*400 = 1960000.
20090727 10:22:19< Dragonking> And this number is.... for one map number of chaces?
20090727 10:22:20< Crab_> then, if we plan 15 turns ahead, 4 bits per 1 tuple, taking symmetry into account, we'll have 0.5Mb of data per map (uncompressed)
20090727 10:22:26< Dragonking> For one map of such size
20090727 10:22:53< Crab_> so, without further tricks, it's in 'not reasonable to ship, quite reasonable to calculate and store' state
20090727 10:23:07-!- Appleman1234 [n=Appleman@131.181.102.53] has joined #wesnoth-dev
20090727 10:23:27< Dragonking> So still calculating this fast is a must
20090727 10:25:33-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20090727 10:25:37< Dragonking> Crab_: Plus by fast calcualtion I mean fast calculation - up to 3-4s on some 1Ghz machine
20090727 10:25:42< Crab_> Dragonking: what if we calculated this only in 'idle time' ?
20090727 10:26:04< Dragonking> Crab_: But we do need thig on demand at the very beggining - I would need - for recruting
20090727 10:26:44< Dragonking> Crab_: Been thinking about something else... about grouping units
20090727 10:26:54< Crab_> Dragonking: but, at the very  beginning, you don't need all this info.
20090727 10:27:20< Crab_> because, at the beginning, unit positions are known
20090727 10:27:32< Dragonking> Well, true
20090727 10:27:34< Crab_> Dragonking: so, the 'result' for the map would take much less space
20090727 10:27:50< Dragonking> Crab_: Well, for beggining there is even something else...
20090727 10:28:04< Dragonking> Crab_: Since usually for beggining there are only leaders
20090727 10:28:13< Dragonking> Crab_: But there is another 'catch' here
20090727 10:28:21-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 10:28:28< Crab_> for 2nd side, not only leader :) - 1st side may have committed to recruiting something
20090727 10:28:43< Dragonking> Crab_: Imagine: we calculate uni 4th sides turn, it's beginning, another 3 already recruited
20090727 10:28:46< Dragonking> Yes
20090727 10:28:59< Crab_> lets see another thing...
20090727 10:29:31-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 10:29:39< Crab_> info about 'for unit type A recruited on hex B, how many turns it would take to reach villages?'
20090727 10:30:30-!- EdB [n=edb@171.117.88-79.rev.gaoland.net] has quit [Remote closed the connection]
20090727 10:30:50< Crab_> let's say 70 villages on map, 10 hexes to recruit, 400 unit types: 28000, divide by 4 because of size/symmetry, 5.6 kb of uncompressed data.
20090727 10:31:08-!- ettin [n=jorda@wesnoth/developer/ettin] has quit [Read error: 60 (Operation timed out)]
20090727 10:31:10< Dragonking> Crab_: But... I don't really see a point of calcualting that
20090727 10:31:33< Dragonking> What it would be used for?
20090727 10:31:36-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090727 10:31:43-!- loonycyborg [n=sergey@79.139.136.93] has joined #wesnoth-dev
20090727 10:31:45< Dragonking> Not always 'villages' are objective on a map
20090727 10:32:52< Crab_> then, what about   'for unit type A (recruited) on hex B, how many turns it would take to reach objective?''
20090727 10:33:41< Dragonking> Ok, sounds better
20090727 10:33:42< Dragonking> :)
20090727 10:34:13< Crab_> I'm thinking about 'precomputing all before wesnoth release, storing that part which is needed on start, then calculating full info in the background and storing it in a cache'
20090727 10:34:36< Crab_> to solve that 'But we do need thig on demand at the very beggining - I would need - for recruting' thing
20090727 10:35:15< Elvenisko> what do you think about scenario with undefined number of turns?
20090727 10:35:29-!- ettin [n=jorda@wesnoth/developer/ettin] has joined #wesnoth-dev
20090727 10:35:30< Dragonking> Crab_: Ok, but still, custom campaigns?
20090727 10:35:55< Dragonking> Crab_: We can't assume we have inefficent calculation and we just pre-cache official things to solve a problem
20090727 10:36:13< euschn> ilor: I currently have two issues with rng: first, I get a load of random number mismatch error when watching any replay. second, it seems that rng is always initialized the same at startup. e.g., if I launch the game and play some moves in a new campaign, the outcome of the fights are always the same
20090727 10:36:24< Elvenisko> i need 25 turns to complete one of my scenarios. how many shall i define in cfg? i can for example add 30 and infinite
20090727 10:37:22< euschn> Ivanovic: I removed this assertion as it is valid for gamestate.snapshot to be empty under certain conditions. Yet, I was unable to reproduce the crash earlier even with the assertion in the code
20090727 10:37:34< Crab_> Dragonking: the same with them: their authors will be able (if they want to make ai smarter) to precompute info which is needed on start, bungle it with the campaign/scenario, and the rest will be done in background
20090727 10:38:25< Ivanovic> euschn: i was using r37234
20090727 10:38:39< Dragonking> Crab_: Well, sounds like a pain for each release of something
20090727 10:40:12< Crab_> Dragonking: why ? it can be a single command-line option to wesnoth. i.e. 'play this game on this map and store start hints to XXX.cfg'. if we have 20 maps in campaign, and 9 seconds per map, then it's 3 minutes of machine time and 0 minutes of human time per release
20090727 10:40:29< Crab_> and it's optional
20090727 10:41:26< Dragonking> Crab_: Not better to somehow integrate it with wesnoth caching? Things can change on the fly, what if map gets modified and someone forgets about updating cache?
20090727 10:42:08< Dragonking> Crab_: Plus 'calculate before release' means that trunk often could have incorrect AI data inside for a mofdified map
20090727 10:42:51< Crab_> well, it all be solved, and it depends on the policy chosen
20090727 10:43:25< Dragonking> TO be honest I'm not really convinced about this
20090727 10:43:26< Crab_> a) calculate on developer side b) calculate as part of release procedures c) calculate on player side
20090727 10:44:03< Crab_> if we go for (c) (the easiest way),
20090727 10:45:10< Crab_> then we'll have choices: c1a) caching this info GLOBALLY after calculation,  c1b) do not caching this info GLOBALLY after calculation.
20090727 10:45:43< Crab_> and c2a) calculating this info on first access c2b) calculating this info in the background and ignoring it on first access if it's not available
20090727 10:46:12-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 10:46:31< Dragonking> Crab_: I'd cache thig for the AI and even store it in a save if needed
20090727 10:46:57< Crab_> note that in campaigns, there's a nice thing called storyscreen
20090727 10:47:07< Dragonking> Yes
20090727 10:47:27< Crab_> if we somehow trick wesnoth into calculating that info during the time player reads the story, even that 9 seconds won't be noticed ;)
20090727 10:48:10< Elvenisko> trying to understand this discussion... still only "trying" :D
20090727 10:48:34< Dragonking> Crab_: Yes, that sounds like a good approach
20090727 10:49:07< Crab_> Elvenisko: it's not a first discussion on this topic, so there's some references to 'this info' without explaining what 'this info' is :)
20090727 10:49:51< Dragonking> Crab_: About floyd-w, we have something that calculates patches, but I don't know which algorithm it uses
20090727 10:50:20< Dragonking> And I did not reall look into how it works, just a glance
20090727 10:50:25< Crab_> Dragonking: current pathfinding uses A* (because it has as a different goal)
20090727 10:50:35< Dragonking> Yes
20090727 10:50:45< Crab_> Dragonking: A* is not useful if we want to have all distances
20090727 10:51:37< Crab_> Dragonking: because the whole point of A* is to use an admissible heuristic to explore only a subset of a graph
20090727 10:51:45< Dragonking> Yes
20090727 10:51:46-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has joined #wesnoth-dev
20090727 10:51:51-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 10:51:57< YogiHH> hello
20090727 10:52:00< Dragonking> hi YogiHH 
20090727 10:52:04< Crab_> hi YogiHH
20090727 10:53:43< Crab_> Dragonking: your current algorithm, it calculates all distances in a single run ? 
20090727 10:54:42< Dragonking> Crab_: After one funcion call I get array of 'scores' for a unit, where score is movement cost to the hex on a map
20090727 10:55:05< Dragonking> Crab_: I call thi for every unit
20090727 10:55:06< Crab_> Dragonking: and that, for one unit, takes 8 s ?
20090727 10:55:10< Dragonking> Crab_: no
20090727 10:55:26< Crab_> Dragonking: 'that, for all units, takes 8 s' ?
20090727 10:56:01< Dragonking> Crab_: Calling, adding allied move-map-scores, thne adding-enemies-move-map, then substracting these two and pushing locations to the vector
20090727 10:56:23< Dragonking> This takes 8s 
20090727 10:57:20-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 10:57:21< Dragonking> hmr
20090727 10:57:22-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 10:57:25< Dragonking> Let me check this again...
20090727 10:59:09< Crab_> it would be interesting to profile this, to see what takes time there...
20090727 10:59:20< Dragonking> Because now I am not sure wheeter I compled wesnoth with debug or not :S
20090727 11:00:11< Dragonking> Crab_: What do you use for profiling?
20090727 11:00:22< Dragonking> I used gprof myself a bit
20090727 11:00:34< Crab_> yes, google perfomance tools
20090727 11:01:58< ilor> euschn:confirmed, strange, will work on it today
20090727 11:01:59-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20090727 11:02:10< euschn> thanks
20090727 11:02:18< Dragonking> Crab_: Do you use something external to interpret results?
20090727 11:02:43< Crab_> I just constructed a graph from the results
20090727 11:02:45< ilor> euschn: the random number mismatch are kind of to be expected in non-combat situations where the seed is not stored, I'll bump the logging down to a warning
20090727 11:02:57< euschn> i see
20090727 11:04:04-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090727 11:08:05-!- Crab_ [i=crab@wesnoth/developer/crab] has quit ["Leaving."]
20090727 11:08:27-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090727 11:15:03< Dragonking> Crab_: haha
20090727 11:15:18< Dragonking> Crab_: ok, I had debug on, so no wonder it was slow :SSSSS
20090727 11:15:43< Crab_> and how much time it takes now ?
20090727 11:15:53< Dragonking> 200ms
20090727 11:16:13< Crab_> that's good)
20090727 11:16:20< Dragonking> Crab_: Sorry to bother you with my stupidity
20090727 11:16:30< euschn> Ivanovic: hm, I was not able to reproduce that assertion failure in r37234, either. Do you have a savegame from before the crash?
20090727 11:17:02< Dragonking> Crab_: Yes, that's really good
20090727 11:17:12-!- ilor_ [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090727 11:23:23< CIA-62> turuk * r37258 /trunk/ (8 files in 7 dirs): Drake Slasher to Arbiter changes.
20090727 11:24:28< CIA-62> turuk * r37259 /trunk/data/core/ (50 files in 2 dirs): Drake Slasher to Arbiter changes.
20090727 11:27:07-!- ilor [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090727 11:28:02-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 11:29:13-!- Elvenisko [n=JA@ig511.imafex.sk] has left #wesnoth-dev []
20090727 11:29:49-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090727 11:31:04-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 11:36:30-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 11:36:49-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090727 11:40:20-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 11:41:20-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Remote closed the connection]
20090727 11:46:49-!- ilor_ [n=user@wesnoth/developer/ilor] has quit [Read error: 110 (Connection timed out)]
20090727 11:49:08< Ivanovic> euschn: no, i don't
20090727 11:49:18< Ivanovic> i just let it run with :droid
20090727 12:00:14-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 12:01:44< CIA-62> turuk * r37260 /trunk/data/ (66 files in 8 dirs): Drake Gladiator to Slasher changes.
20090727 12:03:01-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 12:06:46< Soliton> Turuk_: you really need to get those changes added to wmllint.
20090727 12:07:09< Soliton> i hope you're not doing this by hand.
20090727 12:07:09< Turuk_> Sure, the rule is an easy one to add
20090727 12:07:14< Turuk_> I did
20090727 12:07:25< Turuk_> WMLLINT didn't fix all of the instances where the word appeared
20090727 12:08:02< Soliton> well, why didn't it?
20090727 12:08:09 * Turuk_ shrugs
20090727 12:08:10< Turuk_> Beats me
20090727 12:08:28< Soliton> it's better to make wmllint smarter then to do things ba hand.
20090727 12:08:44< Soliton> well, what for example didn't catch it?
20090727 12:09:04< Soliton> the macro names?
20090727 12:09:14-!- wesbot changed the topic of #wesnoth-dev to: 68 bugs, 239 feature requests, 12 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090727 12:09:52-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 12:10:23< shadowmaster> case sensitiveness?
20090727 12:10:29< Turuk_> shadowmaster: possibly
20090727 12:10:37< Turuk_> I will add in all the variations
20090727 12:10:44< Turuk_> lowercase, capitalized first letter, and all caps
20090727 12:10:59< shadowmaster> wmllint should be case sensitive if it's sane
20090727 12:11:36< Soliton> there was no need to change the macro names so no, wmllint should certainly not try to do that.
20090727 12:11:52< Soliton> so if you did the macro name changes yourself that is fine.
20090727 12:12:03< Turuk_> okay
20090727 12:12:44< Soliton> what about image renaming?
20090727 12:13:20< Soliton> the point of doing the changes with wmllint is to have at least one test case for it to be somewhat sure it'll work for UMC as well.
20090727 12:14:30< Turuk_> I do believe it got the images
20090727 12:15:12< Turuk_> no
20090727 12:15:15< Turuk_> it didn't
20090727 12:15:22< Turuk_> it converted the name in the units file
20090727 12:15:38< Turuk_> but not the actual image file name
20090727 12:15:50< Soliton> right, that's fine.
20090727 12:16:00-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Success]
20090727 12:16:04< zookeeper> "Gladiator to Slasher"?
20090727 12:16:15< zookeeper> oh, right.
20090727 12:16:17< Turuk_> so if I have to change most of this stuff by hand anyways, what's this wmllint thing make easier?
20090727 12:16:25< zookeeper> i guess i'm not keeping up with all this stuff.
20090727 12:16:27< Soliton> ugg
20090727 12:16:51< Soliton> Turuk_: file renames only happen once in mainline...
20090727 12:17:18< Turuk_> Oh?
20090727 12:17:50< Turuk_> let me guess, from one major version to the next?
20090727 12:18:37< Soliton> no, i mean just it happens only in mainline. wmllint is there to update WML anywhere that's a different thing.
20090727 12:18:46< Turuk_> Ah gotcha
20090727 12:19:07< Soliton> the name change is bad anyway though. :-(
20090727 12:19:14< Turuk_> Why?
20090727 12:19:42< Soliton> gladiator -> slasher -> arbiter
20090727 12:19:50< Turuk_> hmm?
20090727 12:20:04< Turuk_> the line goes clasher-slasher-enforcer
20090727 12:20:10< Turuk_> and clasher-arbiter-warden
20090727 12:20:18< Soliton> run it once and it might work right, run it one more time and you have only arbiters.
20090727 12:20:35< Turuk_> oh, for wmllint
20090727 12:20:36< Soliton> i was debating noting that on the forum.
20090727 12:20:52< Soliton> such a change is really bad for automatic conversion.
20090727 12:21:02< Turuk_> well I did it all by hand anyways
20090727 12:21:24< Soliton> i'm not worried about mainline.
20090727 12:22:24< shadowmaster> I thought support for {@} was dropped in 1.7.x?
20090727 12:23:27< Soliton> oh well, if that's how it must be i'd say leave out the gladiator rename from wmllint and make it clear in the release announcment that people need to change that by hand.
20090727 12:24:39< shadowmaster> um... odd; it's still in filesystem.cpp, I was sure it was at least deprecated. nvm
20090727 12:24:44< Soliton> shadowmaster: x=3
20090727 12:25:02-!- Netsplit brown.freenode.net <-> irc.freenode.net quits: Mellar_, corn, allefant, Turuk_, Kenpachi, AI0867
20090727 12:25:04< shadowmaster> then why am I seeing a test for it?
20090727 12:25:37< Soliton> because 1.7.3 is not out yet.
20090727 12:25:41-!- Netsplit over, joins: allefant, Turuk_, Kenpachi, AI0867, corn
20090727 12:25:41< shadowmaster> ah, true :)
20090727 12:26:13< shadowmaster> then the wmllint must be fixed. it's replacing @campaigns with @add-ons
20090727 12:26:18-!- Netsplit over, joins: Mellar_
20090727 12:26:18-!- Appleman1234 [n=Appleman@131.181.102.53] has quit [Read error: 60 (Operation timed out)]
20090727 12:26:19< Soliton> yeah.
20090727 12:29:16< CIA-62> shadowmaster * r37261 /trunk/data/tools/wmllint: 
20090727 12:29:16< CIA-62> wmllint: changed the {@campaigns} replacement with {~add-ons}; it was
20090727 12:29:16< CIA-62> using {@add-ons} instead, which was deprecated and scheduled to be
20090727 12:29:16< CIA-62> removed in 1.7.3.
20090727 12:31:15< CIA-62> turuk * r37262 /trunk/data/tools/wmllint: Updated for the Slasher->Arbiter change.
20090727 12:31:28< CIA-62> turuk * r37263 /trunk/ (changelog players_changelog): Updated changelogs.
20090727 12:32:38< Soliton> Turuk_: please add a note to RELEASE_NOTES about it, too.
20090727 12:32:48< Turuk_> ah Soliton, thanks
20090727 12:33:33< Turuk_> What exactly should I put in there?  That the Gladiator is not in WMLLINT and should be changed by hand?
20090727 12:33:39< Turuk_> There's nothing in there now
20090727 12:34:08< Soliton> everything in there is pasted in the release announcment's Ivanovic does.
20090727 12:34:27< shikadibot> maybe some should poke esr to help with that unit name change
20090727 12:34:28< Soliton> so whatever you'd want to put in those forum posts.
20090727 12:35:14< esr> shikadibot: What unit name needs to be changed?
20090727 12:35:28< shadowmaster> esr: Turuk_ can tell you
20090727 12:35:45< Turuk_> it's already done
20090727 12:36:09< Soliton> Drake Gladiator to Drake Slasher and Drake Slasher to Drake Arbiter
20090727 12:36:24< Soliton> only the first should be done but currently it does too much.
20090727 12:36:59< Soliton> err, only the latter.
20090727 12:38:21< esr> Crap.  That's a non-monotonic change.  I really don't want toto put in in wmllint.
20090727 12:38:26< CIA-62> turuk * r37264 /trunk/RELEASE_NOTES: Updated Release Notes.
20090727 12:39:12< Soliton> esr: yes, it's unfortunate but one change can safely be added.
20090727 12:39:52< esr> Drake Slasher -> Drake Arbiter
20090727 12:52:23-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 12:52:26< Soliton> hmm, still once you converted the gladiator stuff you can't use wmllint again.. maybe it should really just not be added at all. :-/
20090727 13:00:04-!- Elvenisko [n=JA@ig511.imafex.sk] has joined #wesnoth-dev
20090727 13:01:53-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090727 13:07:06< esr> Soliton: That's what I mean by a non-monotonic change.  The right answer is *not to do those * ;-)
20090727 13:07:39< Elvenisko> anyone with some experience with teleports? :)
20090727 13:08:05-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090727 13:08:20< Elvenisko> http://pastebin.com/m40b4001d
20090727 13:08:26< Elvenisko> is this okay?
20090727 13:08:38< Soliton> esr: well, i didn't decide the change. not sure determined people are on that change.
20090727 13:09:46< Soliton> Elvenisko: no, read the macro definition.
20090727 13:11:40< Elvenisko> i need only to add redraw?
20090727 13:12:22< Elvenisko> uff. blinding...
20090727 13:13:24-!- thespaceinvader_ [n=chatzill@cpc3-whit1-0-0-cust305.cdif.cable.ntl.com] has joined #wesnoth-dev
20090727 13:13:46-!- thespaceinvader [n=chatzill@wesnoth/artist/thespaceinvader] has quit [Nick collision from services.]
20090727 13:14:00-!- thespaceinvader_ is now known as thespaceinvader
20090727 13:28:54-!- Elvenisko [n=JA@ig511.imafex.sk] has quit [Read error: 104 (Connection reset by peer)]
20090727 13:59:49-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 14:06:08-!- ilor [n=user@wesnoth/developer/ilor] has joined #wesnoth-dev
20090727 14:07:34-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 14:09:22-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 14:15:47-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 14:27:06-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 14:27:48-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Connection reset by peer]
20090727 14:28:40< Ivanovic> euschn: just tested using head and the assert is gone without any crash or the like as side effect, everything seems to work
20090727 14:29:03< Ivanovic> (okay, no suprise since you removed the assert)
20090727 14:29:14< euschn> yes :)
20090727 14:31:34< Soliton> euschn: https://gna.org/bugs/?14016  might that be related to your changes?
20090727 14:32:00< Ivanovic> thespaceinvader: may i ask you for a favor?
20090727 14:32:26< Ivanovic> thespaceinvader: i just saw that a whole lot of new portraits is added in mainline and i will use the "default ones" for arne and bjarn (the two heros)
20090727 14:32:37< thespaceinvader> Ivanovic: ok
20090727 14:32:39< Ivanovic> because of this the image in the campaign selection is extremely outdated
20090727 14:32:47< euschn> Soliton: yes, its probably related, I will look into it
20090727 14:33:08< thespaceinvader> i think someone was loking into updating those portraits, Ivanovic, but I'm not sure whether there was anything mainline-able made yet
20090727 14:33:18< euschn> Soliton: so you think it is related to randomly generated scenarios?
20090727 14:33:24< thespaceinvader> when we get new artists these days, I tend to point them to campaigns
20090727 14:33:25< Ivanovic> could you create such an image without too much effor using mage-red.png as well as knight.png ?
20090727 14:33:26< Turuk> thespaceinvader: no, they got stuck halfway between
20090727 14:33:35< thespaceinvader> mwhta was the favour, Ivanovic?
20090727 14:33:39< thespaceinvader> *what
20090727 14:33:52< thespaceinvader> dammit, i've got to stop pressing m when i want to type
20090727 14:33:53< Turuk> though at least arne was close: http://www.wesnoth.org/forum/viewtopic.php?p=361425#p361425
20090727 14:34:00< thespaceinvader> stupid multiplayer >.<
20090727 14:34:01< Ivanovic> thespaceinvader: data/campaigns/Two_Brothers/images/campaign_image.png
20090727 14:34:13< Soliton> euschn: yep.
20090727 14:34:47< Ivanovic> making an image like that with the mainline portraits data/core/images/portraits/humans/mage-red.png and data/core/images/portraits/humans/knight.png
20090727 14:36:07< Soliton> euschn: there were several "warning engine: player not found when trying to recall!" messages in the log when it happened.
20090727 14:36:23< Soliton> euschn: so it seems player info is lost somehow.
20090727 14:36:24< euschn> i see
20090727 14:37:33< thespaceinvader> Ivanovic: shouldn't be too much of a problem
20090727 14:37:43< thespaceinvader> would you like black background, or transparent?
20090727 14:37:55< thespaceinvader> and same size?
20090727 14:38:26< Ivanovic> size should probably be the same since it is for the campaign selection dialog
20090727 14:38:49< thespaceinvader> ok
20090727 14:39:06< Ivanovic> not sure if it should be transparent or not
20090727 14:39:15< thespaceinvader> would one of the armoured units whose face you can see be better?
20090727 14:40:58< thespaceinvader> Ivanovic: cos the knight is a very side-on perspective, and doesn't fit in very well with the red mage
20090727 14:43:52< Ivanovic> what about horseman.png ?
20090727 14:44:50< thespaceinvader> just looking, hold on
20090727 14:44:59< thespaceinvader> computer's being slow, virus scanner's running
20090727 14:45:08< Ivanovic> :)
20090727 14:45:52< thespaceinvader> do you want the horse in there was well, or just the rider?
20090727 14:46:58< Ivanovic> hmm, make it just the heads+shoulders of the two, that probably works best
20090727 14:47:10< thespaceinvader> ok, i'll see what i can do
20090727 14:47:51< Ivanovic> though somehow i think that knight and red mage would work, too
20090727 14:48:27< Ivanovic> that is: mirroring red mage and cutting away a little at the buttom of the knight and then just have them side by side merged in an image
20090727 14:52:54< Ivanovic> give me a sec and i will paste you an example (knowing that you can most likely do it a lot better than me)
20090727 14:53:01-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 14:54:35< Ivanovic> thespaceinvader: something like this: http://imagebin.ca/img/8XzDRjOp.png
20090727 14:55:12< Ivanovic> what i basically did is taking the transparent knight plus the transparent red mage (scaled this one down a little because it felt to big when seeing the knight close by)
20090727 14:55:52< thespaceinvader> yeah, i've worked something up briefly hat looks OK to me
20090727 14:55:56< thespaceinvader> i'll PM you
20090727 14:57:26< thespaceinvader> Ivanovic: PMs
20090727 14:58:47< Soliton> https://gna.org/bugs/?14022  what to do with this important bug report?
20090727 14:59:58-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090727 15:06:06< Ivanovic> thespaceinvader: hmm, i am not 100% happy with your version
20090727 15:06:17< Ivanovic> played around a little with it myself using the knight
20090727 15:06:42< thespaceinvader> ok, cool
20090727 15:06:55< thespaceinvader> I can redo it with the knight, if you prefer
20090727 15:06:56-!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"]
20090727 15:08:15-!- stikonas [n=and@ctv-79-132-177-212.vinita.lt] has joined #wesnoth-dev
20090727 15:10:17-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090727 15:11:37< Ivanovic> thespaceinvader: sent you a pm
20090727 15:11:46< Ivanovic> maybe you can do it a little better than I did
20090727 15:11:51< Ivanovic> but i think it works this way
20090727 15:12:31< thespaceinvader> that looks fine to me - I wouldn't do it any differently, to be honest.  You mgiht as well use that one
20090727 15:13:16< Ivanovic> okay, will do it then
20090727 15:13:55< zookeeper> Soliton, well, make the loading faster obviously?
20090727 15:14:24< zookeeper> (or close it)
20090727 15:16:33< zookeeper> AFAIK the long loading times stem from the design that the game re-loads everything, so it's not a pointless complaint IMO. of course one can still say we're not gonna ever rewrite the engine that much to fix it.
20090727 15:18:31< Soliton> zookeeper: well, it's not anywhere that bad here.
20090727 15:19:04< Soliton> and there is no information in the report about what the problem might be other than he uses vista.
20090727 15:19:40< zookeeper> sure, it's not minutes for me either, but they're still too long for my taste.. :p
20090727 15:22:23< YogiHH> zookeeper, Soliton: I'd love to rewrite that part of the engine one day, but as there are not extraordinary loading times atm for normal hardware, i'd say we either request more information or make it "won't do" or "works for me".
20090727 15:23:27< CIA-62> ivanovic * r37265 /trunk/data/campaigns/Two_Brothers/ (8 files in 2 dirs): 
20090727 15:23:27< CIA-62> removed old custom portraits and make use of the mainline portraits instead
20090727 15:23:27< CIA-62> made sure that the right images are used all the times
20090727 15:23:27< CIA-62> create new image for campaign selection based on the mainline knight and red mage portraits
20090727 15:25:24< Ivanovic> thespaceinvader: and thanks for your great work on the drakes, can't wait to see all drakes with portraits and new sprites+animations
20090727 15:25:25< Ivanovic> ;)
20090727 15:25:57< thespaceinvader> no worries
20090727 15:26:02< thespaceinvader> nor can i +D
20090727 15:26:05< thespaceinvader> =D
20090727 15:36:22-!- stikonas_ [n=and@ctv-79-132-177-212.vinita.lt] has joined #wesnoth-dev
20090727 15:39:02-!- mjs-de [n=mjs-de@vpw.wh.uni-dortmund.de] has quit ["On the road again"]
20090727 15:39:20-!- mjs-de [n=mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20090727 15:46:59-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090727 15:49:05< CIA-62> euschn * r37266 /trunk/src/gamestatus.cpp: changed merging of carryover and scenario config (using config::merge_with had unintentional behavior)
20090727 15:50:55< CIA-62> ivanovic * r37267 /trunk/ (changelog players_changelog po/wesnoth-sotbe/de.po): updated German translation
20090727 15:51:02< CIA-62> ivanovic * r37268 /branches/1.6/ (changelog players_changelog po/wesnoth-sotbe/de.po): updated German translation
20090727 15:51:24< Soliton> allefant: addon manager does no delete the old addon when downloading/updating a new one?
20090727 15:51:56< allefant> don't know
20090727 15:53:15< Soliton> can't see it in the get() function.
20090727 15:54:11< Soliton> it should do that so we don't get a mix of versions.
20090727 15:54:24< allefant> let me try add it
20090727 15:54:41< Soliton> thanks.
20090727 15:58:33< CIA-62> elias * r37269 /trunk/data/tools/wesnoth/campaignserver_client.py: [wesnoth_addon_manager] Remove old campaign directory before unpacking a new one.
20090727 15:58:42< allefant> Soliton: can you try if it works?
20090727 15:59:58< Soliton> sure. do you also remove the top level addon.cfg? ;-)
20090727 16:02:27< allefant> not sure, for all i know it might remove "/" right now :P
20090727 16:02:41< allefant> or is addon.cfg an extra file which should be removed?
20090727 16:03:03-!- Kenpachi [n=chatzill@CPE-121-220-195-178.sa.bigpond.net.au] has quit [Read error: 110 (Connection timed out)]
20090727 16:03:08< allefant> i don't have a ~/.wesnoth here myself right now
20090727 16:03:20< allefant> and forgot how to test things myself
20090727 16:03:30< Soliton> addons can have either addon/_main.cfg or addon.cfg
20090727 16:03:39< allefant> oh, i see
20090727 16:03:53< Soliton> you can see it in the tar option further down.
20090727 16:03:59< allefant> won't the addon.cfg ever get deprecated?
20090727 16:04:12< Soliton> maybe.
20090727 16:04:28< Soliton> but some addons might just need that.
20090727 16:04:55< Soliton> i mean just that one cfg file might be enough for simple addons.
20090727 16:05:03-!- Kenpachi [n=chatzill@CPE-121-220-195-178.sa.bigpond.net.au] has joined #wesnoth-dev
20090727 16:05:11< allefant> hm
20090727 16:05:54< allefant> ok, it deletes the whole campaign dir
20090727 16:06:01< allefant> so don't test yet :)
20090727 16:06:33< Soliton> i was about to checkout the diff after your promising comments. :-P
20090727 16:07:59< Soliton> i'm not sure if it's the best idea to put the function into the campaign server client directly.
20090727 16:08:00< allefant> can you give me the command-line to download a test campaign?
20090727 16:09:09< allefant> a small one with many sub-sub-directories, ideally
20090727 16:10:21< Soliton> wesnoth_addon_manager --download=PresetAdvancement -c mydir
20090727 16:10:39< Soliton> it has one sub dir anyway.
20090727 16:11:33< Soliton> Descendants_of_the_Dragon has even more and also a top level cfg.
20090727 16:11:47< CIA-62> mcshark * r37270 /trunk/projectfiles/ (CodeBlocks/wesnoth.cbp CodeBlocks-SCons/wesnoth.cbp): Update according to recent file changes
20090727 16:11:56< allefant> thanks
20090727 16:14:12< Soliton> why do people constantly re-upload completely untested content to the add-on server after i deleted it from there like a million times?
20090727 16:14:40< Soliton> sooner or later i'm not going to bother checking whether they finally updated it and make a cron job to delete them...
20090727 16:15:12< Kenpachi> I am running the latest version of wesnoth from the repos, Synaptic package manager lists the installed version as 1.6.4-2; but when I load the game, it says it's version 1.6.2
20090727 16:15:47< Soliton> which wesnoth
20090727 16:16:43< Soliton> (that's a command you should run and post the output of.)
20090727 16:18:21< Kenpachi> it returns /usr/local/bin/wesnoth
20090727 16:18:49< Soliton> so you're not starting the version from the package manager but a self compiled one.
20090727 16:20:47< CIA-62> elias * r37271 /trunk/data/tools/ (wesnoth/campaignserver_client.py wesnoth_addon_manager): [wesnoth_addon_manager] Really remove campaign. Also remove old-style top-level campaign.cfg.
20090727 16:24:27-!- PK7 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090727 16:26:28< Soliton> allefant: now please backport to 1.6? :-)
20090727 16:28:14 * Soliton can do it too though.
20090727 16:31:31< allefant> svn merge -c 37271 . ../wesnoth-1.6
20090727 16:31:41< allefant> hard part is just waiting for "svn up" to finish :P
20090727 16:32:42< CIA-62> elias * r37272 /branches/1.6/data/tools/wesnoth_addon_manager: backported revision 37271: [wesnoth_addon_manager] Really remove campaign. Also remove old-style top-level campaign.cfg.
20090727 16:33:55-!- Crab_ [i=crab@wesnoth/developer/crab] has quit ["Leaving."]
20090727 16:34:39-!- loonybot [n=loonybot@wesnoth/bot/loonybot] has quit [Read error: 104 (Connection reset by peer)]
20090727 16:37:43-!- Netsplit brown.freenode.net <-> irc.freenode.net quits: MJ999
20090727 16:38:20< ilor> euschn: I've the problem sort of fixed. Sort of, because it uncovered a bad OOS somewhere else and I need to test whether it was there before which will take me some time (read: I'll get back to you tomorrow)
20090727 16:38:51< euschn> ilor: ok, thanks for looking into it
20090727 16:38:51-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 16:38:57-!- Netsplit over, joins: MJ999
20090727 16:39:25-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Connection reset by peer]
20090727 16:39:57-!- stikonas_ is now known as stikonas
20090727 16:40:11< ilor> I could use a faster cpu for this -_-
20090727 16:42:39-!- Netsplit brown.freenode.net <-> irc.freenode.net quits: MJ999
20090727 16:43:14-!- Netsplit over, joins: MJ999
20090727 16:50:06-!- blarumyrran [n=minaise@81-20-159-197.levira.ee] has joined #wesnoth-dev
20090727 16:51:38-!- ardesh_ [n=ardesh@port-92-206-75-230.dynamic.qsc.de] has quit ["Quis custodiet ipsos custodes"]
20090727 16:51:53< CIA-62> soliton * r37273 /branches/1.6/src/unit.cpp: print the config in question when showing the infamous "Attempt to de-serialize a unit with no 'type' field" error to help debugging
20090727 16:53:30< CIA-62> soliton * r37274 /trunk/src/unit.cpp: print the config in question when showing the infamous "Attempt to de-serialize a unit with no 'type' field" error to help debugging
20090727 16:55:12-!- crimson_p [n=irchon@64.201.60.211] has joined #wesnoth-dev
20090727 16:55:41-!- crimson_p [n=irchon@64.201.60.211] has quit [Remote closed the connection]
20090727 16:58:00-!- YogiHH [i=d4ca9d15@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090727 16:59:28-!- crimson_penguin [n=ben@64.201.60.211] has joined #wesnoth-dev
20090727 16:59:39-!- Aethaeryn [n=Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #Wesnoth-dev
20090727 17:00:06-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090727 17:02:53< ilor> ok quick question
20090727 17:03:28< ilor> if I create a scenario with pre-placed units that have traits missing, is MP supposed to work fine on such a scenario?
20090727 17:03:44< ilor> that is, will the generated traits be in sync
20090727 17:04:16< zookeeper> well if the traits are missing then there's nothing that'd be out of sync?
20090727 17:04:27< zookeeper> yes, it should work just fine
20090727 17:04:30< ilor> zookeeper: the missing traits are filled by the engine randomly
20090727 17:05:13< zookeeper> yeah, unless you use random_traits=no...anyway, AFAIK all OOS issues relating to random traits, names and stuff have been fixed
20090727 17:06:03< ilor> I've reverted to r36907, that is, before any of my RNG bad touches, and my simple scenario with just a bunch of units gets out of sync trait-wise
20090727 17:07:04< zookeeper> Soliton, any idea about that?
20090727 17:07:48< Soliton> nope, should be fixed as you said.
20090727 17:08:06< Soliton> what's the scenario?
20090727 17:08:37< ilor> Soliton http://wesnoth.pastebin.com/m2bfae7cf
20090727 17:09:43< ilor> trunk is not a good place to test because there's one RNG f-up there that I haven't commited the fix for yet, and anyway I'm more interested in whether it worked at all before I touched stuff.
20090727 17:11:10< Soliton> does that scenario work with 1.6?
20090727 17:11:26< Soliton> probably easiest to do a binary search for when it broke.
20090727 17:12:03< ilor> yeah, and I need to leave in about 20 minutes ...
20090727 17:12:14< ilor> I'll get to it first thing tomorrow
20090727 17:13:13-!- Polarina [n=polarina@wesnoth/translator/Polarina] has quit [Read error: 60 (Operation timed out)]
20090727 17:13:56< Soliton> zookeeper: type=$theme[$mob_theme].$mob_type means take the [theme] thingy with id=$mod_theme, yes?
20090727 17:14:58< Soliton> or should $mob_theme be a number?
20090727 17:15:28< Soliton> and if so is it 0 or 1-based?
20090727 17:16:31< Soliton> actually it doesn't matter for my problem at hand since the other vars are empty.
20090727 17:16:45-!- Polarina [n=polarina@wesnoth/translator/Polarina] has joined #wesnoth-dev
20090727 17:16:47< zookeeper> Soliton, no, unless someone has implemented something that fancy
20090727 17:17:06< zookeeper> only numbers are valid indices to use
20090727 17:17:20< ilor> Soliton, zookeeper: same deal in 1.6
20090727 17:17:32< Soliton> woah
20090727 17:17:58-!- PK7 [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090727 17:18:40< Soliton> zookeeper: ok 1-based?
20090727 17:18:43< zookeeper> ilor, maybe try putting the units in a prestart event and see if that helps?
20090727 17:18:47< zookeeper> Soliton, 0-based
20090727 17:18:54< Soliton> k
20090727 17:19:31-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090727 17:19:51< Soliton> zookeeper: in that statement what exactly is $theme?
20090727 17:20:22 * ilor wishes his wml-fu was better
20090727 17:20:37< Soliton> i guess $theme alone is not really valid?
20090727 17:20:40< zookeeper> Soliton, nothing. the container is "theme", in that statement you're not referring to "$theme"
20090727 17:20:43< zookeeper> correct
20090727 17:20:56< Soliton> ok, my debugging was a bit misguided then.
20090727 17:21:17< Soliton> but $mob_type alone should be non-empty, yes?
20090727 17:21:36< zookeeper> hmm, yes
20090727 17:22:25< zookeeper> if you try to assign a value or retrieve one from a container itself, might be that it looks for container.__value= or whatever it was then
20090727 17:22:32-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 104 (Connection reset by peer)]
20090727 17:22:47-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 17:22:54< zookeeper> but needless to say i wouldn't advice trying to intentionally use that..
20090727 17:23:04< Soliton> trying to debug Wesband.
20090727 17:23:50< Soliton> could you take a look at Wesband/macros/mob_macros.cfg and tell me if the whole approach is sub-optimal?
20090727 17:24:06< Soliton> it certainly looks fragile to me.
20090727 17:24:25< zookeeper> sure, if you pastebin it
20090727 17:24:35< Soliton> good idea, sec.
20090727 17:24:51< Soliton> http://nopaste.com/p/a6t2ssiaB
20090727 17:26:08< zookeeper> oh...that complex. i don't think i've ever taken a look at wesband before.
20090727 17:26:21< ilor> zookeeper: wrapping the [unit]s in a prestart event synces their traits
20090727 17:27:23< zookeeper> ilor, ok, so it's a problem with stuff in side definitions not getting synced
20090727 17:31:02< zookeeper> Soliton, Giant_Scorpion=Giant Scorpion
20090727 17:31:10< zookeeper> why are those needed?
20090727 17:31:44< zookeeper> something to do with spaces not being allowed in key names, but it sounds like it's probably not necessary
20090727 17:32:24< zookeeper> other than that, it's not very easy to understand that code since i don't know how the details are supposed to work, but doesn't look like there's anything obviously bloaty there
20090727 17:32:29-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090727 17:32:58-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090727 17:34:08< Soliton> hmm, yeah, no idea why the unit ids aren't just directly in the group keys.
20090727 17:34:16< ilor> I'm off, see you tomorrow
20090727 17:34:27-!- ilor [n=user@wesnoth/developer/ilor] has quit []
20090727 17:34:58< Soliton> since those unit keys are very specific anyway and not something that changes from theme to theme.
20090727 17:35:31-!- allefant [n=allefant@allegro/developer/allefant] has quit [Read error: 110 (Connection timed out)]
20090727 17:35:42< CIA-62> euschn * r37275 /trunk/src/playcampaign.cpp: fixed carryover information for randomly generated scenarios, addresses bug #14016
20090727 17:35:59-!- allefant [n=allefant@allegro/developer/allefant] has joined #wesnoth-dev
20090727 17:38:15< euschn> when fixing the above bug, I noticed HttT scenario 16 (Sceptre of fire) occasionally does not place a keep tile on the starting position of the randomly generated map
20090727 17:38:29< euschn> does anyone know what might be the cause?
20090727 17:40:19< zookeeper> not really. around line 110 the tiles around the keep are forced to castle (in case the generator creates a tiny starting castle), but that shouldn't affect the keep.
20090727 18:09:14-!- wesbot changed the topic of #wesnoth-dev to: 67 bugs, 239 feature requests, 11 patches | logs: http://irclogs.wesnoth.org/ | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20090727 18:13:24-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 18:20:19-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090727 18:34:02-!- Noyga [n=lame-z@wesnoth/developer/noyga] has joined #wesnoth-dev
20090727 18:45:19-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090727 18:57:34-!- loonycyborg [n=sergey@wesnoth/developer/loonycyborg] has quit ["KVIrc 3.4.2 Shiny http://www.kvirc.net/"]
20090727 19:01:15-!- euschn [n=chatzill@wesnoth/developer/euschn] has quit [Read error: 110 (Connection timed out)]
20090727 19:22:54-!- boucman [n=rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20090727 19:23:18< boucman> hey all
20090727 19:26:25< Ivanovic> hi boucman
20090727 19:28:01-!- Turuk [n=Turuk@wesnoth/forumsith/turuk] has joined #wesnoth-dev
20090727 19:36:15-!- Turuk_ [n=Turuk@wesnoth/forumsith/turuk] has quit [Read error: 110 (Connection timed out)]
20090727 19:43:08-!- loktor [n=quassel@AAubervilliers-151-1-10-225.w83-114.abo.wanadoo.fr] has joined #wesnoth-dev
20090727 19:59:30-!- Sapient [n=patrickp@wesnoth/developer/sapient] has joined #wesnoth-dev
20090727 20:02:10-!- stikonas [n=and@ctv-79-132-177-212.vinita.lt] has joined #wesnoth-dev
20090727 20:05:04-!- ancestral [n=ancestra@166.205.6.24] has joined #wesnoth-dev
20090727 20:07:53-!- noy [n=Noy@wesnoth/developer/noy] has quit []
20090727 20:09:17-!- silene [n=plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20090727 20:09:23< silene> hi
20090727 20:15:13-!- Kenpachi [n=chatzill@CPE-121-220-195-178.sa.bigpond.net.au] has quit ["ChatZilla 0.9.85 [Firefox 3.0.12/2009070611]"]
20090727 20:18:13< Sapient> yo
20090727 20:19:19< Sapient> AI0867: I think finding two " chars in a line is not a reliable test
20090727 20:19:56< Sapient> quoteparsing is built into the wmliterator class
20090727 20:20:13< Sapient> it would be better to refactor the wmllint code to make use of that
20090727 20:21:04< Sapient> (and to make wmliterator aware of << >>)
20090727 20:25:00-!- grzywacz [n=grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20090727 20:58:06-!- YogiHH [n=chatzill@d097162.adsl.hansenet.de] has joined #wesnoth-dev
20090727 21:13:52-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has joined #wesnoth-dev
20090727 21:23:19-!- ^Noyga^ [n=noyga@AVelizy-151-1-59-98.w82-124.abo.wanadoo.fr] has joined #wesnoth-dev
20090727 21:23:29-!- ^Noyga^ [n=noyga@AVelizy-151-1-59-98.w82-124.abo.wanadoo.fr] has left #wesnoth-dev ["Quitte"]
20090727 21:26:15-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 21:26:33-!- ABCD [n=ABCD@gentoo/contributor/abcd] has quit [Connection reset by peer]
20090727 21:36:19-!- Bob_The_Mighty [n=chatzill@cpc5-brig15-0-0-cust697.3-3.cable.virginmedia.com] has joined #wesnoth-dev
20090727 21:37:28-!- noy [n=Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20090727 21:42:05-!- Crab_ [i=crab@wesnoth/developer/crab] has joined #wesnoth-dev
20090727 21:43:29< boucman> hey Crab_
20090727 21:43:40< Crab_> hi boucman
20090727 21:44:11< boucman> first thing, i'll be away from aug 4 till aug 17
20090727 21:44:28< Crab_> boucman: fully away ?
20090727 21:44:41< boucman> 9->17 i will probably have inet access, but first half I don't know...
20090727 21:45:05< boucman> so you'll have to ask other people if you need help :)
20090727 21:45:11< Crab_> so, 4-9 fully away, then 10-17 'with some inet access' ?
20090727 21:45:19< Crab_> ok
20090727 21:45:19< boucman> yes
20090727 21:46:06< boucman> (sirp being the obvious candidate)
20090727 21:46:11-!- Serei [i=Serei@c-24-118-17-28.hsd1.mn.comcast.net] has joined #wesnoth-dev
20090727 21:46:40< Crab_> ok. most of the things I'll need feedback about are 'policy questions' about fai, so I'll consult with DK more.
20090727 21:47:06< boucman> sounds good
20090727 21:47:25< boucman> and Crab_, I would really like to have your big patch commited before that if possible
20090727 21:48:21< Crab_> boucman: it should be ready soon. i've done all the big things that will make it work without regressions, and now I'm fixing the details.
20090727 21:49:00< boucman> and i started reviewing, which sshould gain me much time
20090727 21:53:09-!- stikonas [n=and@wesnoth/translator/stikonas] has quit [Read error: 110 (Connection timed out)]
20090727 21:54:17< Crab_> I've changed the initialization order (moved engines and aspects to readonly_context). this will require a small  hack to allow fai to work with it (formula_ai inherits from ai_default which needs default_ai_context to initialize. A small hack - a partially uninitialized default_ai_context_proxy.
20090727 21:55:48< Crab_> so, when formula_ai will be first created, it will take the role of a famed formula_ai_library class, and doing any things which use the upper contexts (say, attacking or calculating attacks), will provoke an initialization_exception.
20090727 21:56:34< Crab_> this 'partially initialized' formula_ai will allow us to parse aspects written in fai.
20090727 21:57:24< Crab_> and then, later, formula_ai will be fully initialized. and, in fully initialized state, it will have full access to all those aspects/ other stuff
20090727 21:58:04< Crab_> this can be done more clearly by actually splitting formula_ai into two pieces, but, for now, we (I + DK) decided to keep it as one piece.
20090727 21:58:26< boucman> ok, 
20090727 21:59:01< boucman> Crab_: i'm find with that, I really want the code to converge at this point, we can refine stuff later but we are a bit far of our last base for my taste
20090727 21:59:30-!- loktor [n=quassel@AAubervilliers-151-1-10-225.w83-114.abo.wanadoo.fr] has quit [Remote closed the connection]
20090727 22:00:14< Crab_> boucman: yes, and, because of that there are many things that I code 'in faster, less complex but less elegant way'.
20090727 22:01:04< Crab_> code like http://wesnoth.pastebin.com/m2f317004 , for example
20090727 22:02:05< boucman> Crab_: one (long term) advice though when interacting with open source code (it's a generic advice, but paradoxically, people are less stringent in the professional world)
20090727 22:02:14< boucman> try to do more commits with smaller changes
20090727 22:02:48< boucman> in particular "structural changes" like moving functions around but don't change the actual content creates huge diff which are hell to manage if not commited quickly
20090727 22:03:16< boucman> here you're working on your own so it's no big deal, but a change like the one you are readiying would be hell to merge in a changing codebase
20090727 22:04:42< Crab_> boucman: I know. the problem is that I don't want to break savegames, so I want to commit old-compatible code.
20090727 22:05:10< boucman> and i've repeated that we don't care about savegames in trunk...
20090727 22:05:11< Crab_> boucman: if I'd done this with 'series of smaller commits', then ai config handling in 1.7.3 would be almost certanly broken.
20090727 22:05:39< Crab_> s/1.7.3/1.7.2
20090727 22:05:55< boucman> ok
20090727 22:06:15< Crab_> there are no more config changes pending after this one, so I'll try to plan ahead and keep further commits short.
20090727 22:06:28< boucman> good, thx
20090727 22:07:50< boucman> Crab_: in ai/registry.cpp there are no semantic changes, are they planned for later, or did you forget to remove some stuff ?
20090727 22:09:03< Crab_> boucman: there'll be ~40 new static 'register_aspect_factory<' lines there, but nothing else will be changed there.
20090727 22:09:12< boucman> ok
20090727 22:09:29< boucman> i expected that, wanted to make sure
20090727 22:09:38< Crab_> that example ( http://wesnoth.pastebin.com/m2f317004 ) shows some of them, btw
20090727 22:10:15< boucman> yes, and the commented lines were a hint
20090727 22:15:59-!- blarumyrran [n=minaise@81-20-159-197.levira.ee] has quit []
20090727 22:28:16-!- Espreon [n=espreon@wesnoth/developer/espreon] has joined #wesnoth-dev
20090727 22:47:29-!- Elvish_Pillage2 [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20090727 22:48:33-!- Elvish_Pillager [n=eli@71-10-231-36.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20090727 22:49:32< boucman> Crab_: done proofreading...
20090727 22:49:38< boucman> not much to say, though
20090727 22:49:52< boucman> the structure is quite complicated, but I don't think it can be simplified
20090727 22:50:12< Crab_> ok, feel free to point out strategic errors/ code errors / ugly places / etc
20090727 22:50:42< boucman> it's too cplusplusy for my taste :P
20090727 22:50:58< boucman> but you obviously know your stuff and it's correct, so it's good
20090727 22:51:26-!- PK [n=pk@r74-192-44-206.vctrcmta01.vctatx.tl.dh.suddenlink.net] has quit ["Java user signed off"]
20090727 22:51:30< Sapient> "cplusplusy" ? lol
20090727 22:51:48< Sapient> you are starting to sound like esr now
20090727 22:52:00< esr> ?
20090727 22:52:13< esr> Oh.
20090727 22:52:35< esr> "cpluplusy" = "festering pile of overcomplexity", yes.
20090727 22:52:53< boucman> Sapient: I don't consider them a bad thing, it's just that I've learned C first and it's a totally different language (despite what most people believe)
20090727 22:53:10< boucman> I agree that lot of this stuff (like factories) are workaround for missing language features
20090727 22:58:31< Sapient> if your C code looks very different from your C++ code then that's a good thing
20090727 23:05:51< Sapient> and factories are a common design pattern: http://c2.com/cgi/wiki?CategoryCreationalPatterns
20090727 23:09:17-!- giusef [n=giusef@unaffiliated/giusef] has joined #wesnoth-dev
20090727 23:09:41< Sapient> the lack of common string manipulation/test methods on std::string is kind of annoying though
20090727 23:10:32< Sapient> e.g., a.ends_with(b) ?
20090727 23:10:46< Crab_> Sapient: yes, factories are common design pattern, but when I'm using factories to compensate for missing language features, that's another thing.
20090727 23:12:19< Sapient> what language feature are you simulating?
20090727 23:13:38< Crab_> types as first-class objects, reflection
20090727 23:17:41< Sapient> ah, yeah. those language features are sorely lacking
20090727 23:20:49< CIA-62> thespaceinvader * r37276 /trunk/ (8 files in 5 dirs): Add and wire new Drake Glider portrait, update changelogs, portrait credits.
20090727 23:25:26< zookeeper> Sapient, some people have complained about not being able to add halos to units with {VARIABLE $unit.halo "halo/foo.png"} and the like. any idea about that?
20090727 23:25:33< zookeeper> (that used to work)
20090727 23:27:24< Sapient> that probably doesn't allow direct modifications anymore due to the reconciliation between [object] and direct modification which occurred long ago
20090727 23:27:26< AI0867> I've gotten rid of the symptoms of bug #14018 in r36969, but I'm not sure if there's an underlying issue that should be resolved
20090727 23:27:57< AI0867> Sapient: you're probably right about that, but all I did was stop wmlindent from making my lua code unreadable
20090727 23:28:41< Sapient> zookeeper: if halo attribute should allow direct modification then just submit it as an FR
20090727 23:28:51< Sapient> the relevant code section is unit::read()
20090727 23:29:17< Sapient> the tradeoff is that it won't set the halo correctly anymore when you change the type
20090727 23:29:47< Sapient> and, like all direct modifications, it's lost upon levelup
20090727 23:31:05< Bob_The_Mighty> If you gave a unit a halo to start with, is there any way to 'hide' it until it is needed?
20090727 23:31:06 * zookeeper pokes Bob_The_Mighty
20090727 23:31:13< zookeeper> ah, there he is
20090727 23:31:23< CIA-62> noyga * r37277 /branches/1.6/po/ (wesnoth-lib/fr.po wesnoth-low/fr.po wesnoth-sof/fr.po): French translation update
20090727 23:31:40< zookeeper> well, not as such, but you can give the unit a halo of an empty image
20090727 23:31:41< Bob_The_Mighty> I've been trying in vain for a while to give units halos
20090727 23:31:56< Aethaeryn> Bob_The_Mighty: Galactic Empires?
20090727 23:32:30< Bob_The_Mighty> yes, GE and my rpg era
20090727 23:33:21< Bob_The_Mighty> so if a unit starts with an empty image for a halo, it might work if you just swap the image?
20090727 23:34:05< zookeeper> err, nevermind what i said, i wasn't thinking clearly.
20090727 23:34:26< boucman> Bob_The_Mighty: the easiest way is to drop haloes entirely and use dedicated animations instaed
20090727 23:34:45< zookeeper> boucman, the problem is units that have standing animations already
20090727 23:34:57< boucman> hmm
20090727 23:35:04< boucman> what are you trying to do exactly ?
20090727 23:35:17< zookeeper> in this case i think he needs to be able to put haloes on arbitrary units, not just specific unit types, am i right?
20090727 23:35:22< Sapient> Bob: sounds like it needs a new feature for adding haloes. question is whether it should be via [effect] or via direct modification, or both
20090727 23:35:40-!- noy [n=Noy@wesnoth/developer/noy] has quit [Read error: 60 (Operation timed out)]
20090727 23:35:41< Sapient> should the haloe persist if the unit levels up?
20090727 23:36:04< Bob_The_Mighty> I would think yes
20090727 23:36:09-!- ABCD [n=ABCD@gentoo/contributor/abcd] has joined #wesnoth-dev
20090727 23:36:12< Sapient> then that's EffectWML
20090727 23:36:14< zookeeper> maybe there should be an effect modification that'd apply given direct modifications :x
20090727 23:36:29-!- ABCD_ [n=ABCD@gentoo/contributor/abcd] has quit [Read error: 104 (Connection reset by peer)]
20090727 23:36:37-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20090727 23:36:53 * Sapient tempted to throw something
20090727 23:37:22< Soliton> a cookie?
20090727 23:38:32 * zookeeper has low blood sugar
20090727 23:38:54< Bob_The_Mighty> Sapient, what can you tell me about the bugs of uising variations?
20090727 23:39:01< Bob_The_Mighty> I am allowing the player to switch between variations via a right-click, the thing is the image doesn't update until you click again on the unit (It also seems to be using the attack animations of the previous variation, even though I have specified them in the [variation] tag)
20090727 23:39:19-!- giusef [n=giusef@unaffiliated/giusef] has quit [Read error: 113 (No route to host)]
20090727 23:39:42< Sapient> have you tried refresh?
20090727 23:39:52< zookeeper> :refresh?
20090727 23:39:54< CIA-62> noyga * r37278 /trunk/po/ (wesnoth-lib/fr.po wesnoth-low/fr.po wesnoth-sof/fr.po): French translation update
20090727 23:40:01< AI0867> [redraw][/redraw]?
20090727 23:40:05< Sapient> no, I mean [redraw]
20090727 23:40:17< zookeeper> oh, right
20090727 23:40:30< Bob_The_Mighty> yeah, i tried that
20090727 23:40:42< Sapient> have you tried [redrum] ?
20090727 23:40:58< zookeeper> Sapient, sounds like the same problem as when trying to change unit.type directly, no?
20090727 23:41:26< Sapient> uh, yeah
20090727 23:41:46< Bob_The_Mighty> I've never heard of redrum, does it work on 1.6?
20090727 23:42:02< Sapient> Bob, that was a joke :P  is there an Effect for switching variation?
20090727 23:42:08< zookeeper> there is
20090727 23:42:14< Sapient> ok, then use that
20090727 23:42:35< zookeeper> which is a good point; are you switching variations using an [object] or not?
20090727 23:42:40< Sapient> the effect was added precisely because the direct modification was no longer allowed for that attribute
20090727 23:42:44< Bob_The_Mighty> i am using object
20090727 23:43:09< Sapient> ah, then that would be a bug
20090727 23:43:23< Sapient> unless your variation is not coded correctly in the unit_type
20090727 23:43:29< Sapient> which seems more likely
20090727 23:43:42< Noyga> n8
20090727 23:43:43-!- Noyga [n=lame-z@wesnoth/developer/noyga] has quit ["Quitte"]
20090727 23:43:49< Sapient> is it a mainline unit?
20090727 23:43:55< zookeeper> well, i'm off to bed, have fun
20090727 23:43:56< Bob_The_Mighty> no
20090727 23:44:16< Sapient> ok, try swapping variation of a mainline unit, e.g. Lisar
20090727 23:44:16< Aethaeryn> Only the Runemaster is mainline? Would it be possible to port the whole three lines used by some campaigns? For some reason it doesn't make sense that only one of the three units would be mainline when I could've sworn there were more in default.
20090727 23:44:57< zookeeper> Aethaeryn, yeah, please do, and overwrite the current silly core runemaster
20090727 23:45:02< Sapient> I think the reason you are getting old attack animations is because your variation does not override the merged ones completely
20090727 23:45:03< Bob_The_Mighty> I noticed that the Royal Warrior from TROW is mainline now
20090727 23:45:19< Bob_The_Mighty> Sapient, why would that happen?
20090727 23:45:36< Sapient> a variation is created via a merged config
20090727 23:45:49< zookeeper> of course, porting the SoF runesmith line to core was something i was supposed to do a while back, but...
20090727 23:45:51< Sapient> so anything of the old type that is not overriden, remains
20090727 23:45:56-!- zookeeper [n=l@wesnoth/developer/zookeeper] has quit []
20090727 23:45:57< Aethaeryn> zookeeper: I would if I could. I only have commit access to Wesnoth UMC dev, and I only have official privilages on the wiki. I think I'm in the weirdest state of people.
20090727 23:46:02< Soliton> you can set it to not merge though.
20090727 23:46:44< Aethaeryn> Would make more sense of my wesnoth hostmask said something about the wiki, since that's official and Wesnoth UMC dev is not.
20090727 23:46:48< Soliton> which is what kaleh's variations use because the merge stuff was messy. might have even had bugs.
20090727 23:46:49< Aethaeryn> But meh, left... 
20090727 23:46:54< Bob_The_Mighty> In the wiki it says "inherit: if yes, inherits all the properties of the base unit." I took this to mean, all properties not specified within the variation tag. Am I misguided?
20090727 23:47:33< Soliton> that's right.
20090727 23:47:52< Soliton> it'd be easier if you could just show your WML.
20090727 23:48:49< Bob_The_Mighty> Okay, well here is my unit cfg: http://wesnoth.pastebin.com/d61c43e7a
20090727 23:49:54< Bob_The_Mighty> And here is the event which allows you to swtich between variations: http://wesnoth.pastebin.com/d7e332e39
20090727 23:50:24< Soliton> well, the variation just has two [attack_anim]s.
20090727 23:50:38< Soliton> so the last one is inherited.
20090727 23:51:28< Soliton> and who knows how well the merging inside the anim works out... i'd just not inherit.
20090727 23:51:52< Bob_The_Mighty> whoops, that whole animation chunk was not meant to even be there
20090727 23:52:33< Sapient> Bob_The_Mighty: you should renounce your inheritance
20090727 23:52:40< Sapient> Soliton is right ;)
20090727 23:53:08< Bob_The_Mighty> Oh, but why?
20090727 23:53:50< Soliton> inheritance of tags doesn't work like you think it works.
20090727 23:54:13-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20090727 23:54:57< Bob_The_Mighty> What happens to a) objects and b) variable modifications when a variation is switched?
20090727 23:55:56< Soliton> apply_to=variation is like leveling.
20090727 23:57:00-!- YogiHH [n=chatzill@wesnoth/developer/yogihh] has left #wesnoth-dev []
20090727 23:57:22< Bob_The_Mighty> can the variation take new name?
20090727 23:58:09< Soliton> of course.
--- Log closed Tue Jul 28 00:00:30 2009