--- Log opened Mon Dec 12 00:00:32 2011
20111212 00:03:04-!- Gallaecio [~quassel@84.120.216.157.dyn.user.ono.com] has quit [Remote host closed the connection]
20111212 00:04:49-!- wesbot changed the topic of #wesnoth-dev to: String/feature-freeze active for trunk | 163 bugs, 328 feature requests, 13 patches | Logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20111212 00:06:06< anonymissimus> Soliton: no poblems visible to me atm
20111212 00:08:42< Soliton> cool, if you also checked with #19161 please mark that report accordingly.
20111212 00:09:03< anonymissimus> sure I want to understand what caused that bug
20111212 00:09:07-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20111212 00:09:47< anonymissimus> although you're no longer calling handle_turn()...
20111212 00:10:29< Soliton> of course.
20111212 00:11:22< Espreon> AI0867: Thanks.
20111212 00:11:32< anonymissimus> is the init_side added to the replay save file of a scenario from the bug
20111212 00:12:27< Soliton> anyone knows how FD_CLOEXEC works?
20111212 00:13:41< Soliton> i'm setting that on sockets created by wesnothd and it restarts via system() but the socket FDs don't get closed.
20111212 00:14:46< Soliton> (and system() calls exec() or similar in the end i believe.)
20111212 00:19:14< Soliton> i guess the other solution is to close all FDs above 2 which i've seen a couple of programs do. not particularly elegant but maybe more portable.
20111212 00:20:29< loonycyborg> Soliton: What do you want to accomplish anyway?
20111212 00:21:35< Soliton> loonycyborg: the server is leaking FDs on restart.
20111212 00:22:24< Soliton> graceful restart that is, when it closes the server socket and waits for all games to finish.
20111212 00:23:24< loonycyborg> Does it call itself with system() or something?
20111212 00:24:17< Soliton> yes.
20111212 00:25:18< loonycyborg> I'd that that it already is hack, so closing all fds > 2 isn't so bad.
20111212 00:25:52< Soliton> how else would you do the restart?
20111212 00:26:21-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20111212 00:27:02< loonycyborg> Non-hackish solution would be either having whatever scripts that start the server handle this or have the server reinit itself without running system() on itself.
20111212 00:28:12< Soliton> first solution is hackish as well. how would the second one work?
20111212 00:28:37< loonycyborg> Actually if you want to upgrade the binary the second solution doesn't make sense..
20111212 00:30:02-!- chris_oelmueller is now known as chrisoelmueller
20111212 00:30:23< Soliton> well, the system() call is on the run script not the actual running binary.
20111212 00:31:28< loonycyborg> Honestly, I'm not sure I understand the purpose of graceful restart.
20111212 00:32:44< Soliton> to avoid ending every game and kicking everyone when restarting...
20111212 00:33:57< loonycyborg> How usually do you perform a graceful restart?
20111212 00:34:44< Soliton> currently whenever the server is about to run out of memory... :-P
20111212 00:35:29-!- vultraz [~chatzilla@124.109.10.221] has quit [Ping timeout: 252 seconds]
20111212 00:36:22< loonycyborg> If the next wesnothd isn't spawned by orginial wesnothd that was terminated then I don't see how fds can leak.
20111212 00:36:49< loonycyborg> Shouldn't they all be cleaned up once all processes using them are terminated?
20111212 00:37:01< Soliton> sure.
20111212 00:37:38< Soliton> as mentioned before there is that system() call on restart...
20111212 00:39:49-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20111212 00:40:59< Soliton> anyhow, i guess brute force closing also has the benefit of closing other leaking FDs like the command fifo and others i might be forgetting.
20111212 00:41:11-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20111212 00:43:46< loonycyborg> Soliton: I think system() might be a red herring here.
20111212 00:43:50-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 248 seconds]
20111212 00:44:17< Soliton> not sure what you mean.
20111212 00:45:38< loonycyborg> It doesn't matter. wesnothd doesn't have an opportunity to pass fds to child processes because it doesn't start them in this case.
20111212 00:45:49-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 00:46:10< loonycyborg> Probably it just leaks them and that's all.
20111212 00:46:39< Soliton> i don't know what case you mean but in the case of a graceful restart i'm talking about it surely does.
20111212 00:47:49< Soliton> and afaic it is quite logical since it itself calls system() thus passing on FDs by default.
20111212 00:48:10< loonycyborg> utils/mp-server/run_server seems to always restart it based on wesnothd's return code.
20111212 00:48:26< loonycyborg> Does the same thing happen in case of graceful restart?
20111212 00:48:56< Soliton> of course and the return code is... 0.
20111212 00:49:47< Soliton> unless something bad happens after the graceful restart in which case there'd be a bit of a problem which the run script should catch though.
20111212 00:50:02< loonycyborg> And it wouldn't pass fds to the run script because it's a parent, not a child.
20111212 00:50:51< Soliton> sigh, wesnothd calls the run script...
20111212 00:51:43< Soliton> i'm not talking about hypothetical issues anyway. it's a real problem i can reproduce and which caused major issue with wesnothd on at least one occasion.
20111212 00:52:04< Soliton> if we could get back to an actual solution that'd be more useful.
20111212 00:57:04< Soliton> you can also check it right now on wesnoth.org with a command like: netstat -plant | grep 14998.*CLOSE_WAIT | wc -l
20111212 00:57:52< Soliton> those do not happen to be 100+ users trying to disconnect there since days...
20111212 00:58:20< loonycyborg> CLOSE_WAIT? That's familiar..
20111212 00:59:52< loonycyborg> If you close an accepting socket all connections will be in this state for some time.
20111212 01:00:46-!- vultraz [~chatzilla@124.109.10.221] has joined #wesnoth-dev
20111212 01:01:25< Soliton> for some time in this case being until the process dies..
20111212 01:01:51< Soliton> anyhow before you go further with speculations i suggest you check out the graceful restart feature if you don't believe me.
20111212 01:02:15< loonycyborg> btw do you use SO_REUSEADDR?
20111212 01:04:42< Soliton> on the server socket? SDL does that, i think.
20111212 01:04:55< Soliton> restart would obviously not work well otherwise.
20111212 01:07:05< loonycyborg> std::system(restart_command.c_str()) <- I do believe you :P
20111212 01:08:46-!- Sytyi [~Sytyi@252-156-133-95.pool.ukrtel.net] has quit [Quit: Leaving.]
20111212 01:14:05< loonycyborg> Soliton: So the problem is that old wesnothd instance doesn't bother to close those connections?
20111212 01:14:47< Soliton> since that'd defeat the purpose of a graceful restart, yes...
20111212 01:15:41< Soliton> we want people to stay connected and be able to finish their games.
20111212 01:16:11< loonycyborg> No. If it'll only close those in CLOSE_WAIT state it won't :P
20111212 01:16:58< Soliton> who's going to be in that state?
20111212 01:17:05< loonycyborg> Soliton: And old instance will automatically terminate once everyone have left?
20111212 01:17:13< Soliton> yes.
20111212 01:17:19< Soliton> once all games ended.
20111212 01:19:53< loonycyborg> Anyway, netstat -plant doesn't work for me on the server since only root can use -p
20111212 01:20:27< Soliton> well, leave out -p then...
20111212 01:20:58< Soliton> it wasn't relevant to the example command at all.
20111212 01:22:39-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 268 seconds]
20111212 01:24:00-!- vultraz [~chatzilla@124.109.10.221] has quit [Read error: Connection reset by peer]
20111212 01:28:19-!- vultraz [~chatzilla@124.109.10.221] has joined #wesnoth-dev
20111212 01:29:45-!- Kolbur [~die@89.204.130.68] has left #wesnoth-dev []
20111212 01:30:53< loonycyborg> Soliton: Actually I wonder to which process those sockets belnong :P
20111212 01:30:58< loonycyborg> *belong
20111212 01:33:00-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 01:36:02-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20111212 01:38:26< Soliton> hmm, actually it seems i put the FD manipulation code at the wrong place.
20111212 01:40:01< Soliton> where it only affects client side sockets.
20111212 01:40:40< loonycyborg> You mean FD_CLOEXEC code?
20111212 01:41:16< Soliton> yep, works now.
20111212 01:41:37< loonycyborg> Great! :)
20111212 01:42:48< Soliton> i'm guessing the command fifo already gets closed properly since its FD is not actually leaked.
20111212 01:43:50< Soliton> ah, indeed it is on restart.
20111212 01:44:19< loonycyborg> An alternative way to implement graceful restart would be to have wesnothd stop accepting connections and somehow instruct its parent run script to run another server.
20111212 01:44:53< Soliton> yeah, hacky though IMO.
20111212 01:45:16< Soliton> you'd have to touch some file or so and it'd likely not be instant restart.
20111212 01:45:32< loonycyborg> Maybe send a signal instead?
20111212 01:45:50< loonycyborg> afaik bash scripts can set handlers..
20111212 01:47:15< Soliton> yeah, probably the only way since the script does a wait and it doesn't seem like you can make that timed so you can check other stuff inbetween.
20111212 01:48:09< Soliton> probably even less portable then though.
20111212 01:53:17< loonycyborg> Probably as portable as bash :P
20111212 01:56:14< shadowmaster> dash can also use trap to set signal handlers, so I presume that feature is found in most bourne shell derivatives
20111212 01:59:07< loonycyborg> signals are part of C standard, so probably even windows implements them in some way.
20111212 02:00:23-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 252 seconds]
20111212 02:00:27< AI0867> Invalid WML found: Window builder: parsing resolution tip with empty 'id' field. Will become mandatory in 1.9.7. (2)
20111212 02:00:32< loonycyborg> anyway, gn
20111212 02:00:32< AI0867> mordante: ^
20111212 02:03:49-!- Atomicbomb [~quassel@61.94.49.108] has joined #wesnoth-dev
20111212 02:04:00-!- Blueblaze [~Blueblaze@99.4.147.142] has quit [Quit: Blueblaze]
20111212 02:06:30-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 255 seconds]
20111212 02:08:00-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 02:18:37< shadowmaster> loonycyborg: Windows doesn't implement them at the system level I think, and the Microsft CRT only supports (emulates?) a small subset of signals
20111212 02:19:52< shadowmaster> I had completely forgotten that wesnothd needs to run on Windows, so I was thinking only about Linux and BSD derivative systems
20111212 02:20:48-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 02:24:53-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20111212 02:25:16-!- PolarPanda [~quassel@unaffiliated/peterporty] has quit [Read error: Connection reset by peer]
20111212 02:25:30< loonycyborg> shadowmaster: Yes, but graceful restart would most likely be ever used on unices..
20111212 02:25:49-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20111212 02:26:37< CIA-85> ai0867 * r52244 /trunk/ (19 files in 9 dirs): (log message trimmed)
20111212 02:26:37< CIA-85> Merge and update r49342 from 1.8 branch
20111212 02:26:37< CIA-85> rework MP chat log as a GUI2 dialog with colored display of messages
20111212 02:26:37< CIA-85> Conflicts:
20111212 02:26:37< CIA-85> changelog
20111212 02:26:38< CIA-85> players_changelog
20111212 02:26:38< CIA-85> projectfiles/CodeBlocks-SCons/wesnoth.cbp
20111212 02:26:43-!- anonymissimus_ [~chatzilla@HSI-KBW-078-042-163-105.hsi3.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20111212 02:26:47< CIA-85> ai0867 * r52245 /trunk/src/gui/dialogs/chat_log.cpp: 
20111212 02:26:47< CIA-85> Forward port and update r49343 from 1.8:
20111212 02:26:47< CIA-85> fix a TODO with slider in chat_log GUI2 dialog, change per-page count to 1000, and correctly disable the 'next' and 'previous' buttons when pages are not available
20111212 02:26:57< CIA-85> ai0867 * r52246 /trunk/data/gui/default/window/chat_log.cfg: 
20111212 02:26:57< CIA-85> Forward port r49471 from 1.8
20111212 02:26:57< CIA-85> reindent chat_log.cfg and fix a harmless error message in the logs
20111212 02:27:09< CIA-85> ai0867 * r52247 /trunk/data/gui/default/window/chat_log.cfg: 
20111212 02:27:09< CIA-85> Forward port r49472 from 1.8
20111212 02:27:09< CIA-85> add borders to chat log dialog to make it look nicer
20111212 02:27:23< CIA-85> ai0867 * r52248 /trunk/data/gui/default/window/chat_log.cfg: 
20111212 02:27:23< CIA-85> Forward port r49473 from 1.8
20111212 02:27:23< CIA-85> wrap messages in chat log dialog in a scroll label
20111212 02:27:33< CIA-85> ai0867 * r52249 /trunk/src/gui/dialogs/chat_log.cpp: 
20111212 02:27:33< CIA-85> Forward port r49474 from 1.8
20111212 02:27:33< CIA-85> in chat log window, try to escape some chars which might break pango markup.
20111212 02:27:46< AI0867> wesbot: tag 1.8.6
20111212 02:27:59< AI0867> !tag 1.8.6
20111212 02:28:00< shikadibot> AI0867: Tag 1.8.6, revision 49488 (ivanovic) on Thu May 12 11:41:55 2011:
20111212 02:28:03< shikadibot> AI0867: tagged 1.8.6 (using r49486)
20111212 02:28:04< shikadibot> AI0867: Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=49488
20111212 02:28:06-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-105.hsi3.kabel-badenwuerttemberg.de] has quit [Ping timeout: 255 seconds]
20111212 02:28:10-!- anonymissimus_ is now known as anonymissimus
20111212 02:28:36< AI0867> shadowmaster, mordante: porting is done, though I think some tweaks may still be needed
20111212 02:29:12< AI0867> the COUNT_PER_PAGE constant for one
20111212 02:29:49< AI0867> also, the scrollbox's horizontal size (it can shrink to way smaller than the dialog)
20111212 02:29:50< CIA-85> espreon * r52250 /trunk/ (6 files in 3 dirs): Ran umcpropfix.
20111212 02:30:17< AI0867> and the slider's behaviour (it may need some hysteresis added or something)
20111212 02:33:08-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20111212 02:36:13-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 02:54:49-!- Vorpal [~AnMaster@unaffiliated/anmaster] has quit [Ping timeout: 252 seconds]
20111212 02:56:58-!- Atomicbomb [~quassel@61.94.49.108] has quit [Read error: Connection reset by peer]
20111212 03:01:56-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20111212 03:06:22-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-105.hsi3.kabel-badenwuerttemberg.de] has quit [Quit: done building targets]
20111212 03:07:36-!- Atomicbomb [~quassel@125.160.85.235] has joined #wesnoth-dev
20111212 03:12:25-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 03:28:41-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20111212 03:47:19-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20111212 03:55:20-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has quit [Ping timeout: 252 seconds]
20111212 03:58:59-!- PolarPanda is now known as Jhubbz_
20111212 04:02:00< CIA-85> alink * r52252 /trunk/src/scripting/lua.cpp: 
20111212 04:02:00< CIA-85> Make Lua get_time_of_day() handles off-map correctly and consistently
20111212 04:02:00< CIA-85> (same way regardless of the illumination parameter, and also as all the rest of the game does)
20111212 04:02:05< CIA-85> alink * r52253 /trunk/src/ (7 files in 2 dirs): 
20111212 04:02:05< CIA-85> Clean get_time_of_day function names:
20111212 04:02:05< CIA-85> get_time_of_day() = global ToD
20111212 04:02:05< CIA-85> get_time_of_day(loc) = local ToD using ToD areas
20111212 04:02:06< CIA-85> get_illuminated_time_of_day(loc) = local ToD using ToD areas and illumination
20111212 04:02:06< CIA-85> alink * r52251 /trunk/src/display.cpp: 
20111212 04:02:07< CIA-85> Remove never reached code randomly added by the messy r50620.
20111212 04:02:07< CIA-85> Because tod_manager is only enabled in game but then game_display override this.
20111212 04:02:08< CIA-85> alink * r52254 /trunk/src/tod_manager.cpp: Move code around for better readability. No logic change
20111212 04:02:10< CIA-85> alink * r52255 /trunk/src/tod_manager.cpp: 
20111212 04:02:10< CIA-85> Remove a misplaced and always true validation
20111212 04:02:11< CIA-85> (by construction, see parse_times)
20111212 04:02:24-!- Jhubbz_ [~quassel@unaffiliated/peterporty] has quit [Remote host closed the connection]
20111212 04:02:46-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20111212 04:03:52< alink> anonymissimus: ^should be simpler now (see r52253 msg)
20111212 04:04:16< alink> BTW no logic change
20111212 04:16:36-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20111212 04:19:58-!- horon [~horon@nttkyo316066.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has joined #wesnoth-dev
20111212 04:50:49-!- Ivanovic_ [~ivanovic@dtmd-4d0bf909.pool.mediaWays.net] has joined #wesnoth-dev
20111212 04:50:49-!- Ivanovic_ [~ivanovic@dtmd-4d0bf909.pool.mediaWays.net] has quit [Changing host]
20111212 04:50:49-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20111212 04:54:11-!- Ivanovic [~ivanovic@dtmd-4db22775.pool.mediaWays.net] has quit [Ping timeout: 252 seconds]
20111212 04:54:51-!- Ivanovic_ is now known as Ivanovic
20111212 05:54:00-!- Blueblaze [~Blueblaze@adsl-99-4-147-142.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20111212 06:17:19-!- Atomicbomb is now known as michellebot
20111212 06:18:39-!- michellebot is now known as Atomicbomb
20111212 06:19:14-!- Gambit [~quassel@wesnoth/developer/grickit] has quit [Read error: Connection reset by peer]
20111212 06:27:38-!- Aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 252 seconds]
20111212 06:48:30-!- csarmi [csarmi@178-164-141-152.pool.digikabel.hu] has quit []
20111212 06:53:31-!- PolarPanda [~quassel@unaffiliated/peterporty] has quit [Read error: Connection reset by peer]
20111212 06:57:57-!- Aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 07:06:12-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Read error: Connection reset by peer]
20111212 07:06:29-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 07:13:20-!- Atomicbomb [~quassel@125.160.85.235] has quit [Remote host closed the connection]
20111212 07:16:05-!- Atomicbomb [~quassel@125.160.85.235] has joined #wesnoth-dev
20111212 08:11:20-!- Kolbur [~die@89.204.155.201] has joined #wesnoth-dev
20111212 08:25:26-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20111212 08:31:20< Soliton> loonycyborg: actually i think currently graceful restart should work on windows as well. how's the network rewrite coming, btw?
20111212 08:37:10-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20111212 09:03:11-!- Appleman1234 [~Appleman1@ppp59-167-222-56.qld.adsl.internode.on.net] has joined #wesnoth-dev
20111212 09:30:13-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20111212 09:31:02-!- Blueblaze [~Blueblaze@adsl-99-4-147-142.dsl.hstntx.sbcglobal.net] has left #wesnoth-dev []
20111212 09:33:33-!- Atomicbomb is now known as michellebot
20111212 09:34:19-!- michellebot is now known as Atomicbomb
20111212 09:39:38-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 252 seconds]
20111212 09:48:22-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20111212 10:01:25-!- Sytyi [~Sytyi@93-236-133-95.pool.ukrtel.net] has joined #wesnoth-dev
20111212 10:10:12-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20111212 10:16:51-!- Vorpal [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20111212 10:31:18-!- mjs-de [~mjs-de@d120154.adsl.hansenet.de] has joined #wesnoth-dev
20111212 10:35:28< Ivanovic> moin
20111212 10:51:23-!- Atomicbomb [~quassel@125.160.85.235] has quit [Remote host closed the connection]
20111212 11:12:27-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20111212 11:22:24-!- Sytyi [~Sytyi@93-236-133-95.pool.ukrtel.net] has quit [Remote host closed the connection]
20111212 11:27:59-!- sytyi [~sytyi@93-236-133-95.pool.ukrtel.net] has joined #wesnoth-dev
20111212 11:35:35-!- Atomicbomb [~quassel@125.161.20.107] has joined #wesnoth-dev
20111212 11:38:11-!- Atomicbomb [~quassel@125.161.20.107] has quit [Remote host closed the connection]
20111212 11:41:24-!- Atomicbomb [~quassel@125.161.20.107] has joined #wesnoth-dev
20111212 11:58:02-!- Octalot [~noct@host86-186-159-239.range86-186.btcentralplus.com] has joined #wesnoth-dev
20111212 11:59:53-!- loonybot [~loonybot@46.138.81.125] has joined #wesnoth-dev
20111212 11:59:53-!- loonybot [~loonybot@46.138.81.125] has quit [Changing host]
20111212 11:59:53-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20111212 12:07:40< loonycyborg> Soliton: wrt network rewrite for now I've managed to make it accept connections and handle login stuff using asio.
20111212 12:08:46< loonycyborg> asio's way of having everything in callbacks is quite different from how wesnothd currently works.
20111212 12:10:17< Soliton> hmm, i see.
20111212 12:47:58-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20111212 12:50:39-!- stikonas [~gentoo@ctv-213-164-110-84.vinita.lt] has joined #wesnoth-dev
20111212 12:50:39-!- stikonas [~gentoo@ctv-213-164-110-84.vinita.lt] has quit [Changing host]
20111212 12:50:39-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20111212 13:09:59-!- Atomicbomb [~quassel@125.161.20.107] has quit [Remote host closed the connection]
20111212 13:16:51-!- Atomicbomb [~quassel@125.161.20.107] has joined #wesnoth-dev
20111212 13:35:16-!- Gambit [~quassel@wesnoth/developer/grickit] has joined #wesnoth-dev
20111212 13:54:37-!- Atomicbomb [~quassel@125.161.20.107] has quit [Read error: Connection reset by peer]
20111212 13:55:06-!- Atomicbomb [~quassel@125.161.20.107] has joined #wesnoth-dev
20111212 14:17:37-!- negusnyul [~negusnyul@1F2EA95B.dsl.pool.telekom.hu] has joined #wesnoth-dev
20111212 14:29:03-!- PolarPanda [~quassel@unaffiliated/peterporty] has joined #wesnoth-dev
20111212 14:44:47-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20111212 15:11:21-!- sytyi [~sytyi@93-236-133-95.pool.ukrtel.net] has quit [Quit: Leaving.]
20111212 15:24:07-!- Cookiee [~quassel@unaffiliated/cookiee] has joined #wesnoth-dev
20111212 15:25:46-!- Elvish_Pillager [~eli@dhip-149.coburn.residences.colby.edu] has joined #wesnoth-dev
20111212 15:42:19-!- horon [~horon@nttkyo316066.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has quit [Quit: Leaving...]
20111212 15:42:24-!- Atomicbomb [~quassel@125.161.20.107] has quit [Remote host closed the connection]
20111212 15:48:21-!- Crendgrim [~crend@77-22-113-8-dynip.superkabel.de] has joined #wesnoth-dev
20111212 16:09:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20111212 16:16:00-!- Cookiee [~quassel@unaffiliated/cookiee] has quit [Remote host closed the connection]
20111212 16:32:19-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20111212 16:33:36-!- Sytyi [~Sytyi@93-236-133-95.pool.ukrtel.net] has joined #wesnoth-dev
20111212 17:17:05-!- negusnyul_ [~negusnyul@145.236.69.239] has joined #wesnoth-dev
20111212 17:17:19-!- negusnyul [~negusnyul@1F2EA95B.dsl.pool.telekom.hu] has quit [Ping timeout: 252 seconds]
20111212 17:26:48-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20111212 17:36:34-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 17:48:01-!- Upthorn [~ogmar@c-98-208-38-181.hsd1.ca.comcast.net] has quit [Ping timeout: 252 seconds]
20111212 17:48:36< shadowmaster> Hm, Boost 1.48 entered Debian testing.
20111212 18:00:08-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 18:03:59< CIA-85> shadowmaster * r52256 /trunk/data/campaigns/Liberty/scenarios/03_A_Strategy_Of_Hope.cfg: Liberty scenario 3: make Relana start facing in Baldras' direction
20111212 18:04:49-!- wesbot changed the topic of #wesnoth-dev to: String/feature-freeze active for trunk | 162 bugs, 328 feature requests, 13 patches | Logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20111212 18:12:37-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20111212 18:21:41< CIA-85> shadowmaster * r52257 /trunk/src/addon/manager.cpp: 
20111212 18:21:41< CIA-85> Revert r46885 in order to fix bug #18983
20111212 18:21:41< CIA-85> Conflicts:
20111212 18:21:41< CIA-85> src/addon/manager.cpp
20111212 18:24:55< shadowmaster> Ivanovic: do you think an exception could be made in order to fix https://gna.org/bugs/?18969 ?
20111212 18:28:34< Ivanovic> shadowmaster: uhm, what exactly is the case where this note: is used?
20111212 18:28:43< Ivanovic> do you have a screenshot of "is state"?
20111212 18:28:48< shadowmaster> http://shadowm.rewound.net/f/tutorial-note.png
20111212 18:29:01< shadowmaster> took me 4 minutes to upload
20111212 18:29:34< Ivanovic> honestly, it does not look bad or wrong to me, so i'd say: no, lets keep the current state
20111212 18:29:57< Ivanovic> AI0867: i saw that you committed some stuff last night, is this the forward port of the chat dialog in 1.8.x?
20111212 18:29:59< shadowmaster> I don't see any good reason to let a style issue like this make it to 1.10 if it's possible to fix it without rewriting the whole thing.
20111212 18:30:36< Ivanovic> and: is the forward port completed now (so that i can/should run a pot-update and inform the translators)
20111212 18:30:37< shadowmaster> Unless, of course, you don't care about style. I think we left that direction around 1.2.
20111212 18:32:33-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 245 seconds]
20111212 18:34:32< vultraz> shadowmaster: um...maybe I'm stupid...but how is removing a word in a message considered breaking the feature freeze?
20111212 18:35:29< Gambit> string freeze
20111212 18:35:32< shadowmaster> vultraz: the string needs to be retranslated afterwards because the word in question is part of the same string
20111212 18:35:42< shadowmaster> that breaks what Gambit said
20111212 18:35:58< vultraz> oh..
20111212 18:40:26-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20111212 18:49:33< CIA-85> shadowmaster * r52258 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: Fix "Invalid key 'maximumvaluelabel='" error
20111212 18:51:52< CIA-85> shadowmaster * r52259 /trunk/data/gui/default/window/chat_log.cfg: 
20111212 18:51:52< CIA-85> gui2/tchat_log: Add tooltip/helptip class ids
20111212 18:51:52< CIA-85> This removes the following error:
20111212 18:51:52< CIA-85> Invalid WML found: Window builder: parsing resolution tip with empty
20111212 18:51:52< CIA-85> 'id' field. Will become mandatory in 1.9.7. (2)
20111212 18:52:16< shadowmaster> AI0867: ^
20111212 18:54:47< CIA-85> shadowmaster * r52260 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: GUI2 window definition descriptions aren't supposed to be translatable yet
20111212 18:55:01-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20111212 18:55:17< CIA-85> shadowmaster * r52261 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: Add a colon to a slider label for consistency with other GUI1 and GUI2 dialogs
20111212 18:57:40< CIA-85> shadowmaster * r52262 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: Empty strings can't be marked as translatable!
20111212 18:58:05-!- mjs-de [~mjs-de@d120154.adsl.hansenet.de] has quit [Remote host closed the connection]
20111212 18:59:39< CIA-85> shadowmaster * r52263 /branches/1.8/data/gui/default/window/chat_log.cfg: 
20111212 18:59:39< CIA-85> gui2/tchat_log: GUI2 window definition descriptions aren't supposed to be translatable yet
20111212 18:59:39< CIA-85> (Backported from trunk, r52260.)
20111212 19:03:10< CIA-85> shadowmaster * r52264 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: Set size constraints to match the campaign menu, and add standard border size
20111212 19:04:20< CIA-85> shadowmaster * r52265 /trunk/data/gui/default/window/chat_log.cfg: gui2/tchat_log: Dialog titles should use Title Case
20111212 19:12:55-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20111212 19:18:56< shadowmaster> the C++ for this dialog reeks of overengineering
20111212 19:25:15< shadowmaster> that's hardly the only problem, though. The dialog size may vary wildly depending on the individual pages' contents, rendering the slider unusable; the button locations will then change
20111212 19:26:33< shadowmaster> perhaps we should restrain the dialog size to a fixed value (800x480 would be a good candidate since it's the minimum resolution)
20111212 19:29:18< shadowmaster> or use a multipage widget to hold the log pages, although I don't think that will scale well for large chat logs (not that the current approach scales well either)
20111212 19:31:47< CIA-85> shadowmaster * r52266 /trunk/ (2 files in 2 dirs): 
20111212 19:31:47< CIA-85> gui2/tchat_log: More minor cosmetic changes
20111212 19:31:47< CIA-85> * Get rid of a useless column with spacers and the Close button; the
20111212 19:31:47< CIA-85> latter has been moved to the previous column and its horizontal
20111212 19:31:47< CIA-85> alignment changed accordingly.
20111212 19:31:48< CIA-85> * Disable the pages slider when the page count is 1.
20111212 19:34:26< shadowmaster> wesbot: seen Alarantalara
20111212 19:34:26< wesbot> shadowmaster: The person with the nick Alarantalara last spoke 5d 18h ago.  5d 15h ago person left:
20111212 19:36:46< shadowmaster> Alarantalara: may want to check https://gna.org/bugs/?19179 if you didn't get an email about it
20111212 19:53:42-!- negusnyul__ [~negusnyul@dsl5401B259.pool.t-online.hu] has joined #wesnoth-dev
20111212 19:56:47-!- negusnyul_ [~negusnyul@145.236.69.239] has quit [Ping timeout: 240 seconds]
20111212 19:56:55-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has joined #wesnoth-dev
20111212 20:03:33-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20111212 20:07:31-!- grzywacz [~grzywacz@89-72-193-138.dynamic.chello.pl] has joined #wesnoth-dev
20111212 20:07:31-!- grzywacz [~grzywacz@89-72-193-138.dynamic.chello.pl] has quit [Changing host]
20111212 20:07:31-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20111212 20:09:46< shadowmaster> alink: I have found a nice rendering engine benchmark map in mainline!
20111212 20:10:13< alink> shadowmaster: i am listening
20111212 20:10:16< shadowmaster> the 1v1 Thousand Strings Garrison keeps CPU usage around 55% for me when using animated terrains
20111212 20:10:27< alink> checking
20111212 20:11:00-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20111212 20:11:00-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20111212 20:11:00-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20111212 20:11:16< mordante> seruvs
20111212 20:11:31< alink> ih mordante
20111212 20:11:36< mordante> hi alink 
20111212 20:12:31< mordante> AI0867, will have a look later, nice to see the chat dialog has been ported
20111212 20:13:09< alink> shadowmaster: not so horrible here, but indeed
20111212 20:13:50< alink> shadowmaster: can you perceive an animation in ford? because I can't but :sunset tell me there is one
20111212 20:14:13< shadowmaster> yes, fords are currently animated 
20111212 20:14:32< shadowmaster> it's ridiculously subtle
20111212 20:14:40< alink> ah i almost see it when puting my nose on the screen, very subtle
20111212 20:16:20< alink> and yet, seems animated at high speed
20111212 20:16:30< alink> at least it is not multi-hex
20111212 20:17:12< mordante> AI0867, nevermind I see shadowmaster already fixed the warnings
20111212 20:18:19< shadowmaster> I've been running with the theory that eleazar's monitor has an unusual contrast setting
20111212 20:19:39< alink> it seem to be blue animated water with some almost opaque brown ford image on top of it, masking most of the animated parts
20111212 20:20:24< alink> and :layers tell me there is 50 layers there, but, as often, most of them transparent
20111212 20:22:30< alink> shadowmaster: ok thanks I will try to investigate that when I finish my ToD optimization 
20111212 20:22:48< shadowmaster> by transparent you mean they only contain pixels with the alpha channel set to 0 (fully transparent)?
20111212 20:23:23< alink> yes seems so, at least from what I can see in :layers
20111212 20:23:58< alink> it's a common result of the hex-cutting done by some WML terrain rules
20111212 20:24:13< shadowmaster> sounds wasteful
20111212 20:24:56< alink> yes and no, SDL RLE compression should reduce that to 0 pixels to blit
20111212 20:25:50< alink> but the layers itself are still handled by the code
20111212 20:26:45< alink> and it's specially bad for ToD coloring, zooming etc because we uncompress, do complex work on transparent images and the recompress
20111212 20:27:01< alink> but after that it's cached
20111212 20:27:17< alink> s/the/then
20111212 20:29:21< alink> I had ideas how to filter them but need to evaluate the relative cost of the scanning needed to spot them, and when do such step exactly
20111212 20:30:43< alink> I think I even had a prototype. it didn't really help the framerate but I forget to check for the other operations mentioned above 
20111212 20:32:34< alink> anyway the main problem with transparent images is when they are animated. That hurts framerate for visually nothing and was a common error in the paste
20111212 20:33:14< alink> that's why I don't like seeing 50 layers above ford
20111212 20:35:12< alink> (if 2 layers are animated but out-of-sync, the hex is refreshed 2 times more often)
20111212 20:39:54-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-105.hsi3.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20111212 20:43:52< alink> shadowmaster: all these scorpions (animated because of the fire under them) on castles (so overlap the 3 hexes above) also cause a 4 hexes invalidation for each one
20111212 20:44:35< alink> so, even if small petrified arachnids, it's like having an army of big animated units
20111212 20:50:20< Kolbur> hehehe, crazy Doc was at work there ^^
20111212 20:50:31< anonymissimus> btw is the pango font size modification with span size=small etc supposed to work in [label] ?
20111212 20:51:08< anonymissimus> since it didnt recently for me; and I'm sure the syntax was right since I copied it form somewhere else where it worked
20111212 20:53:41< anonymissimus> color modification in the same was works btw
20111212 20:53:43< alink> anonymissimus: I think I vaguely remember something about disabling them for team label to avoid abuse, not sure if it apply here too
20111212 20:53:56< alink> oh ok nvm then
20111212 20:54:32< alink> maybe "span size=8" ?
20111212 20:54:43< anonymissimus> what sense does it make to disable them for player-set labels ?
20111212 20:54:56< anonymissimus> I tried a lot nothing worked
20111212 20:55:22< anonymissimus> I dont suppose some player actually knows or would enter <span ...> ... ;)
20111212 20:55:47-!- grzywa [~grzywacz@89-72-193-138.dynamic.chello.pl] has joined #wesnoth-dev
20111212 20:55:51< alink> well they are colored using the team-color, if you can change that, you can pretend it's a label from another team or something
20111212 20:55:51-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Read error: Connection reset by peer]
20111212 20:56:02< anonymissimus> the two usages of the label structure are unfortunately linked
20111212 20:56:30< alink> anyway, as I said, only a vague memory about it
20111212 20:56:47< anonymissimus> it still would be visible only to the right team
20111212 20:56:57< Soliton> yes, color changing must be disabled.
20111212 20:57:02< anonymissimus> if its a team-specific label
20111212 20:57:18< alink> yes but yo can add a public label disguised as a team label for the other team
20111212 20:57:24< Soliton> if you can still allow other modifications that might be fine.
20111212 20:59:09 * anonymissimus would use up his complete remaining time for sure upon getting the syntax for the incorrectly colored label fine
20111212 21:01:01< shadowmaster> I don't think it'd be a good idea to let people set labels to larger font sizes either
20111212 21:03:45< Gambit> Unless team labels had some look about them that can't be replicated with public ones
20111212 21:03:46< Soliton> it'd be fine to distinguish between player and WML set labels in any case IMO.
20111212 21:04:13< Gambit> (Like putting images in front of them. Different ones for public, team, and WML)
20111212 21:08:43< anonymissimus> all; i just checked; I can set a colored label via action wml and pango, the exact same pango doesn't work when set via the right click menu however
20111212 21:08:47-!- Octalot [~noct@host86-186-159-239.range86-186.btcentralplus.com] has quit [Ping timeout: 240 seconds]
20111212 21:08:52< alink> maybe such icon could indicate the exact hex associated to the label (not always obvious)
20111212 21:08:57< anonymissimus> so it seems fixed
20111212 21:09:19-!- Upth [~ogmar@public-nat1.scc.losrios.edu] has joined #wesnoth-dev
20111212 21:09:20-!- Upth is now known as Upthorn
20111212 21:10:46< anonymissimus> shadowmaster: I certainly wish to be able to change font size in labels; have a perfect usecase for it
20111212 21:11:07< anonymissimus> via wml that is
20111212 21:11:18< anonymissimus> not via the menu of course
20111212 21:12:25< shadowmaster> which is?
20111212 21:13:32-!- grzywa [~grzywacz@89-72-193-138.dynamic.chello.pl] has quit [Quit: :wq]
20111212 21:17:05-!- Sytyi [~Sytyi@93-236-133-95.pool.ukrtel.net] has quit [Quit: Leaving.]
20111212 21:18:50< anonymissimus> a substitution for the number terrains in ma addon Settlers of Wesnoth
20111212 21:18:59< anonymissimus> using labels
20111212 21:20:00-!- crimson_pingvin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20111212 21:20:01-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Read error: Connection reset by peer]
20111212 21:20:01-!- crimson_pingvin is now known as crimson_penguin
20111212 21:20:50< anonymissimus> they look like so http://imagebin.org/188252
20111212 21:21:33< anonymissimus> and since [terrain_type] cannot be transferred over the network I now use labels instead, it would certainly be nice to have the numbers larger than the ***s
20111212 21:24:26< anonymissimus> actually, I should say that binary data can't be transferred over the network such as these images
20111212 21:24:41< anonymissimus> since then I could still use [item] instead
20111212 21:27:06< shadowmaster> mordante: do you remember the number of that bug in the editor that I wanted you to check?
20111212 21:27:47< shadowmaster> ah, #18928, nvm
20111212 21:28:04< shadowmaster> I guess I'll test with Boost 1.48
20111212 21:31:36< shadowmaster> hm, big download. maybe another day :) (or as soon as aptitude is rebuilt against 1.48)
20111212 21:44:42-!- Octalot [~noct@host109-145-74-254.range109-145.btcentralplus.com] has joined #wesnoth-dev
20111212 21:44:59-!- Octalot [~noct@host109-145-74-254.range109-145.btcentralplus.com] has quit [Client Quit]
20111212 21:45:12-!- Octalot [~noct@host109-145-74-254.range109-145.btcentralplus.com] has joined #wesnoth-dev
20111212 21:50:58< mordante> shadowmaster, wesnoth compiles with boost 1.48 for me, just need a minor patch
20111212 21:52:15< mordante> shadowmaster, replace the boost foreach.hpp with this file http://paste.debian.net/149025/
20111212 21:53:51< shadowmaster> mordante: is that the 1.48 Boost.foreach?
20111212 21:54:46< mordante> jup
20111212 22:03:15< shadowmaster> I seem to be unable to compile using that
20111212 22:04:18-!- Crendgrim [~crend@77-22-113-8-dynip.superkabel.de] has quit [Quit: see ya]
20111212 22:11:42-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Read error: Connection reset by peer]
20111212 22:29:24-!- crimson_penguin [~ben@184.71.170.250] has joined #wesnoth-dev
20111212 22:29:24-!- crimson_penguin [~ben@184.71.170.250] has quit [Changing host]
20111212 22:29:24-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20111212 22:33:26< anonymissimus> shadowmaster: the triggering point of bug #19177 seems to be in unlawful orders
20111212 22:33:51< anonymissimus> the reporter attached a wrong start-of-scenario save
20111212 22:33:55< shadowm_laptop> the entry subject doesn't seem to indicate so
20111212 22:34:05< anonymissimus> but I also cant reproduce it
20111212 22:34:12< shadowm_laptop> it says "A Strategy of Hope"
20111212 22:34:43< anonymissimus> in unlawful orders there's the "switch between loyalists and undead" code
20111212 22:35:01< anonymissimus> that looks suspicious for triggering such bug
20111212 22:35:28< shadowm_laptop> it should be marked as Need Info since the report is clearly not very well redacted
20111212 22:35:47< shadowm_laptop> that is, assuming your suspicion is correct; otherwise, Works for Me is good
20111212 22:37:30< anonymissimus> the code wasn't changed since long though
20111212 22:38:30-!- Appleman1234 [~Appleman1@ppp59-167-222-56.qld.adsl.internode.on.net] has quit [Read error: Connection reset by peer]
20111212 22:39:33 * anonymissimus feels uncomfortable about marking something "works for me" since its means I don't understand why it works for me but not everyone
20111212 22:46:57-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20111212 22:51:59-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 252 seconds]
20111212 23:00:27< mordante> I'm off night
20111212 23:00:43-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20111212 23:02:06< anonymissimus> Soliton: you sure don't want to mark bug #16299 "ready for test" ? :)
20111212 23:02:32< anonymissimus> think I understand the cause of bug #19161
20111212 23:11:38-!- mjs-de [~mjs-de@wh.uni-dortmund.de] has quit [Remote host closed the connection]
20111212 23:12:02< Kolbur> hello, there is someone who tries to register to the forum but his confirmation email doesn't get send
20111212 23:16:17< shadowm_laptop> he should email forums@wesnoth.org unless he can reach me on IRC
20111212 23:18:23-!- negusnyul__ [~negusnyul@dsl5401B259.pool.t-online.hu] has quit [Read error: Connection reset by peer]
20111212 23:19:23< shadowmaster> that is, me
20111212 23:19:34< shadowmaster> not shadowm_laptop, which is less reliable
20111212 23:20:57-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: log archiving in progress]
20111212 23:23:27< Kolbur> ok
20111212 23:37:12-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20111212 23:48:34-!- mjs-de [~mjs-de@d120154.adsl.hansenet.de] has joined #wesnoth-dev
--- Log closed Tue Dec 13 00:00:49 2011