--- Log opened Fri Sep 24 00:00:25 2010
20100924 00:01:38< alink> Ivanovic: Pentium dual-core (2M Cache, 2.60GHz, 800MHz FSB), GeForce 9500 GT (nvidia driver), debian stable
20100924 00:02:13< alink> and mega fan on the GPU ;)
20100924 00:02:45< Ivanovic> :)
20100924 00:03:58< alink> and yes my specific OpenGL hack is probably slower on older or cheaper GPU (at least it was like that few years ago)
20100924 00:04:01-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100924 00:04:57< Ivanovic> i am using a phenom 2 x6 1060 (2.8GHz max clockspeed), radeon 3850 (aka rv670) with the open source radeon driver and mesa git
20100924 00:11:31< alink> hmmm, and it scale badly with terrain complexity, which is consistent with my suspicion about transparent terrain images
20100924 00:21:02-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100924 00:21:25< CIA-35> zookeeper * r46675 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/02_Peaceful_Valley.cfg: Colour-shifted Darken Volk's hair and robes a bit.
20100924 00:21:47< Ivanovic> alink: what scales badly? current mode or your opengl hack?
20100924 00:21:58< alink> opengl hack
20100924 00:23:39< alink> because we don't have SDL RLE compression skipping all our dummy terrain images (I think)
20100924 00:24:08< alink> but I will optimize that soon (for SDL trunk, but that should help the OGL hack too) 
20100924 00:24:31< Ivanovic> no surprise for me that it does not scale well since you are using the sdl technics for opengl which is meant to be used differently for best performance
20100924 00:26:13< alink> this dummy images thing is more: bigger weight for the current flaws of our code
20100924 00:27:04< alink> even with SDL it cost us some memory and loading, for example we zoom, ToD color them etc.. and it's all empty images
20100924 00:27:14< esr> fabi: You have twoo different definitions of MULTImlscope is barfing on this; fix it, please?PLAYER_GOLD in LoW.  w
20100924 00:27:24< esr> Aaargh.
20100924 00:27:33< esr> That got mangled.
20100924 00:28:09< esr> fabP You have two different definitions of MULTIPLAYER_GOLD in LoW.  Fix it, please?
20100924 00:29:28< CIA-35> esr * r46676 /trunk/data/ (6 files in 5 dirs): Clear up a bunch of minor string FIXMes from before 1.8.
20100924 00:30:17< Ivanovic> esr: bad one!
20100924 00:30:28< esr> Ivanovic: ?
20100924 00:30:30< Ivanovic> stringfreeze because of release!
20100924 00:30:42< esr> Huh?  
20100924 00:30:53< Ivanovic> ah, sorry, was a trunk commit, not 1.8
20100924 00:30:54< Mica> esr: How are things?
20100924 00:31:01-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]]
20100924 00:31:25< esr> Mica: OK. What's on your mind?
20100924 00:31:36< Mica> Nothing, it's just been a long time since we talked, hasn't it?
20100924 00:31:50< esr> Yes.
20100924 00:32:02< Mica> How've you been?
20100924 00:32:14< esr> Busy, as usual.
20100924 00:32:26< shadowmaster> who isn't
20100924 00:32:33< Mica> Me! XD
20100924 00:32:45< shadowmaster> fabi: pling
20100924 00:33:11< Mica> brb
20100924 00:33:58< shadowmaster> fabi: you seem to have tried to contact me by PM in multiple occasions lately, which is a bad idea since I rarely pay attention to query windows' existence in my main session. May I know what's it about?
20100924 00:34:04< shadowmaster> *IRC PM
20100924 00:34:39< shadowmaster> ("hello" isn't the best way to attract my attention either)
20100924 00:35:28< shadowmaster> and I'm afk, so if you will, hilight me in a public channel so my laptop session reports on my KDE notifications area
20100924 00:45:51-!- iwaim [~iwaim@ns1.alib.jp] has quit [Remote host closed the connection]
20100924 00:47:16-!- iwaim [~iwaim@ns1.alib.jp] has joined #wesnoth-dev
20100924 01:13:36-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100924 01:17:32-!- anonymissimus [~chatzilla@HSI-KBW-095-208-007-247.hsi5.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819]]
20100924 01:24:18-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 240 seconds]
20100924 01:36:49-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20100924 01:38:25-!- Blueblaze [~Blueblaze@adsl-99-171-161-94.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100924 01:46:47-!- Upth [ogmar@adsl-75-26-200-109.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100924 01:46:47-!- Upth is now known as Upthorn
20100924 01:52:24-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100924 01:58:09-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100924 02:01:41-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Quit: Leaving]
20100924 02:02:56-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20100924 02:04:13< CIA-35> esr * r46677 /trunk/data/core/units/ (8 files in 3 dirs): Remove some unused cruftv that's been hanging around since 1.2.
20100924 02:07:24-!- TheBuzzSaw [~buzz@75-174-86-121.bois.qwest.net] has joined #wesnoth-dev
20100924 02:08:10-!- Mica_ [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20100924 02:11:08-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Ping timeout: 276 seconds]
20100924 02:11:22-!- Mica_ is now known as Mica
20100924 02:11:46-!- PeterPorty [~Pete@pc-241-115-214-201.cm.vtr.net] has joined #wesnoth-dev
20100924 02:12:12-!- Upth [~ogmar@adsl-75-26-198-81.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100924 02:13:03-!- Upthorn [ogmar@adsl-75-26-200-109.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 245 seconds]
20100924 02:15:37-!- wesbot changed the topic of #wesnoth-dev to: 159 bugs, 300 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100924 02:25:07< CIA-35> esr * r46678 /trunk/data/core/units/elves/Archer.cfg: Remove small but fatal impediment....
20100924 02:26:37< CIA-35> esr * r46679 /trunk/data/core/units/humans/Mage.cfg: Typo fix.
20100924 02:26:42-!- TheBuzzSaw [~buzz@75-174-86-121.bois.qwest.net] has left #wesnoth-dev []
20100924 02:27:22-!- PeterPorty [~Pete@pc-241-115-214-201.cm.vtr.net] has quit [Quit: Saliendo]
20100924 02:34:05< Mica> esr: I'm having a strange problem with connecting to Wesnoth's official server.
20100924 02:34:20< Mica> I think it has to do with my specific Wesnoth client. Can you help?
20100924 02:34:49< esr> Alas, probably not - I don't grok multiplayer well.  But I'll try.
20100924 02:35:11-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 02:35:16< Mica> Of course now it's working.
20100924 02:35:37< Mica> Thanks anyways xD
20100924 02:58:04-!- Khezar [~Greywhind@harkness-112.resnet.brown.edu] has joined #wesnoth-dev
20100924 02:58:04-!- Khezar [~Greywhind@harkness-112.resnet.brown.edu] has quit [Client Quit]
20100924 03:03:45-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100924 03:05:08< Mica> esr: Is there a list of 1.4's terrain codes?
20100924 03:05:23< Mica> or any way I can find out what they were.
20100924 03:05:49< esr> Mica: Not any more.  You'd have to look at old versions of the wiki, I think.  Why do you need to know?
20100924 03:06:20< Mica> esr: There's a map from 1.4 that won't load in the current editor, and I need to convert it. How can I look at old versions of the wiki?
20100924 03:06:38< Gambit> Mica: wmlint
20100924 03:06:48< Mica> I SUCK WITH WMLINT D:
20100924 03:06:56< esr> Oh, that's a much easier problem.  Get wmllint from the 1.6 branch.
20100924 03:07:09< Mica> I can't use wmllint. I .. just can't.
20100924 03:07:41< Gambit> Mica: get the wml ide timotei made
20100924 03:07:44< Gambit> two clicks
20100924 03:07:48< esr> Mica: Oh, don't snivel. Look at the cannded invocations in data/tools/Makefile.
20100924 03:07:57< Mica> No I mean I seriously can't.
20100924 03:08:03< Mica> It doesn't work for me.
20100924 03:08:11< Mica> Someone walked me through it and it wouldn't work
20100924 03:08:24< esr> Then ur don it wrong.  Find out how to do it right.
20100924 03:08:30 * shadowm_laptop giggles while watching esr deal with a normal user.
20100924 03:08:35< billynux> Mica, <stupid positive message> Don't underestimate wht you can do. </msg>
20100924 03:08:43< Mica> Lol
20100924 03:09:32-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100924 03:10:54< esr> Mica: Clue - I *never* invoke wmllint by hand.  I use thw makefile productions, which do all the hairy options for you.
20100924 03:11:36< Mica> How do I use the makefile ?
20100924 03:13:26< esr> Find the production that matches what you want to do (they all have explanatory comments) and give the production name as anb argument to the make command.  
20100924 03:15:21< esr> For example, what I normally do is just type "make", which the makefile interprets as "male sanity-check" because that's the first production in the file, whisch depends on the "unresolved" and "lint" productions.  So it runs those.
20100924 03:15:38< Mica> o.o
20100924 03:15:58< esr> Each one calls a tool with a bunch of optioms I set up once and then forgot about.
20100924 03:16:10< esr> s/male/make/
20100924 03:16:44< Mica> Why do I need 1.6's wmllint? Doesn't 1.8 come with it?
20100924 03:16:45< esr> Or you can use the GUI.  Two clicks, as Gambit said.
20100924 03:17:01< esr> Acctually...
20100924 03:17:25< esr> The 1.8 lint may work. You said it was a 1.4 map?
20100924 03:17:34< Mica> Yeah
20100924 03:17:43< Mica> How do I begin to run it? Like how do I open it up?
20100924 03:17:47< Mica> Open the command line?
20100924 03:18:00< esr> That's OK, then.  The trunk version of wmllint should do.
20100924 03:18:47< esr> Mica: You're on a Windows machine, Linux, Mac?  What?
20100924 03:18:56< Mica> Windows
20100924 03:20:06< esr> Shit.  Well, that means you need to install Python first....and this is where I tell you to talk to a Windows dev, as I know bery little about Windows and am constantly working at knowing even less..
20100924 03:20:31 * shadowm_laptop is rolling on the floor laughing.
20100924 03:20:43< Mica> I hate windows, and if I had the know how, I'd run linux.
20100924 03:20:45< Mica> But I don't.
20100924 03:20:47 * billynux doesn't have non free software installed
20100924 03:21:04< billynux> Mica, Ubuntu, also two clicks away
20100924 03:21:18< Mica> For me, everything's a little more complicated.
20100924 03:21:26< shadowm_laptop> I advise you stay away from Linux if you aren't tech-savvy
20100924 03:21:30< Mica> ^
20100924 03:21:33< Mica> Exactly.
20100924 03:21:36< Mica> I can figure some things out
20100924 03:21:38< shadowm_laptop> there's really no such a thing like a user-friendly Linux.
20100924 03:21:39< esr> shadowmaster: You know Windows.  Stop laughing and start helping, you worthless antipodean :-)
20100924 03:21:42< Mica> But my main problem is, I fear code.
20100924 03:21:44< shadowm_laptop> Ubuntu is a pathetic attempt at that.
20100924 03:21:58< shadowm_laptop> esr: sorry, I'd not dare run Python on Windows.
20100924 03:22:03< shadowm_laptop> I barely dare to run it on Linux already.
20100924 03:22:13< billynux> well, I disagree, but won't get into argument about it
20100924 03:22:29< Mica> shadowm_laptop: Then how do YOU run wmllint? You can, can't you?
20100924 03:22:44< shadowm_laptop> I can but I have no idea how to do it properly.
20100924 03:22:51 * shadowm_laptop ignores anything said above about Makefiles.
20100924 03:23:04< esr> *You* have no excuse.
20100924 03:23:06< shadowm_laptop> at least, not how to do it without draining CPU time
20100924 03:23:19< shadowm_laptop> since it needs to parse mainline for some reason.
20100924 03:23:42< esr> That's to build a cross-reference table.  Necessary, sorry.
20100924 03:23:43< shadowm_laptop> esr: yes, I have one, and it's that it's against my principles to run wmllint, you know that.
20100924 03:24:24< esr> Clearly your principles require some adjustment. :-)
20100924 03:24:37< shadowmaster> 22:23:48 <shadowmaster> I don't care
20100924 03:24:50< Mica> Eh. It's okay.
20100924 03:24:58< Mica> I can just try to look up the old wiki and update manually.
20100924 03:25:03< Mica> It actually doesn't bother me..
20100924 03:25:22< shadowmaster> but it's still amusing to see your biased reaction to Windows users ;)
20100924 03:25:39< shadowmaster> considering we're already in the 21st century, not in the 90s.
20100924 03:30:19< Mica> shadowmaster: Can you help me get an old version of the wiki?
20100924 03:30:29< shadowmaster> how so?
20100924 03:30:53-!- shadowm_laptop is now known as NotShadowmaster
20100924 03:30:59< Mica> There has to be a way to load up a version of the wiki.
20100924 03:31:00< NotShadowmaster> Mica: just give me the map and I'll run wmllint on it for you.
20100924 03:31:12< Mica> Ah. Ok. I'll PM it to you.
20100924 03:31:14< NotShadowmaster> of course there's a way to do what you want, it's called "History" :)
20100924 03:31:18 * NotShadowmaster looks at shadowmaster 
20100924 03:31:53< Mica> Submitted.
20100924 03:32:13< NotShadowmaster> er-
20100924 03:32:14< NotShadowmaster> Undead Empire?
20100924 03:32:24< NotShadowmaster> isn't that part of the Undead Empire add-on?
20100924 03:32:25< Mica> Yep
20100924 03:32:35< NotShadowmaster> I'm pretty sure there's a 1.8 version already
20100924 03:32:36< Mica> jb and Mythological have not updated it
20100924 03:32:42< NotShadowmaster> oh
20100924 03:32:43< Mica> We're talking on Wesnoth now
20100924 03:33:01< NotShadowmaster> what else would be talking about?
20100924 03:33:24 * NotShadowmaster stares in awe at an amusing SDL bug.
20100924 03:33:31< Mica> SDL ?
20100924 03:33:36< NotShadowmaster> shikadibot: SDL?
20100924 03:33:36< shikadibot> SDL: Simple Direct-Media Layer/library
20100924 03:33:46< Mica> Ah
20100924 03:34:14< NotShadowmaster> Mica: I think there's a 1.6 version of UE
20100924 03:34:22< Mica> jb said 1.4
20100924 03:34:50< NotShadowmaster> there's one in wesnoth-umc-dev at least
20100924 03:35:10< Mica> Mythological says no, and he uploaded that, I think he might have put it in the wrong spot
20100924 03:35:28< NotShadowmaster> are you talking to them right now?
20100924 03:35:32< Mica> Yeah
20100924 03:35:36< NotShadowmaster> how?
20100924 03:35:51< Mica> Wesnoth, we're playing a  game together and while I wait for my turn I'm doing this :p
20100924 03:36:11< NotShadowmaster> shadowmaster: go stalk them
20100924 03:36:42< shadowmaster> I can't.
20100924 03:36:54< shadowmaster> I don't have any add-ns installed. ;(
20100924 03:37:20< Mica> lol
20100924 03:38:29< shadowmaster> and it seems I got the wrong add-on
20100924 03:39:00< Mica> I think it's the very last one, if sorted by names
20100924 03:39:53< shadowmaster> is it the Zombie Ap. RPG?
20100924 03:40:57< Mica> I don't think so. I dled that and the other Zombie Apoc. thing though, just to be saf
20100924 03:41:01< Mica> safe*
20100924 03:42:08-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 03:42:10< shadowmaster> getting it from the web server is so much faster than trying a direct download from the add-ons server.
20100924 03:42:27< shadowmaster> 300 KB/second compared to 2 KB/second
20100924 03:43:38< shadowmaster> Mica: it  was the RPG add-on
20100924 03:43:46< shadowmaster> not the plain Zombie APocalypse one
20100924 03:44:14< shadowmaster> it looks cool, btw.
20100924 03:44:23 * shadowmaster hasn't played any MP add-ons ever.
20100924 03:45:36< Mica> Lol I dunno
20100924 03:45:38< Mica> It's OK
20100924 03:45:42< Mica> This is my first time playing it
20100924 03:46:35< CIA-35> ai0867 * r46680 /trunk/src/log.hpp: include sstream from log.hpp, as it is needed to use lg::wml_error
20100924 03:52:44-!- Mythological [Ilccov@77.28.89.98] has joined #wesnoth-dev
20100924 03:54:18-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100924 03:54:26-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 265 seconds]
20100924 03:54:29< Mica> Hey Mythological 
20100924 03:54:47-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100924 04:01:33-!- NotShadowmaster is now known as shadowm_laptop
20100924 04:07:56-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]]
20100924 04:10:57-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20100924 04:15:10-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100924 04:17:18-!- gabba [~gabba@wesnoth/developer/gabba] has quit [Read error: Connection reset by peer]
20100924 04:18:00-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100924 04:18:55< gabba> hey boucman, I saw you were looking for me. Nope I'm not dead, just very busy with school. I hope to look this weekend at the bugs that have been piling up, and solve at least one.
20100924 04:21:16-!- gabba [~gabba@wesnoth/developer/gabba] has quit [Read error: Connection reset by peer]
20100924 04:43:54-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100924 04:57:43-!- Ivanovic_ [~ivanovic@dtmd-4db2b956.pool.mediaWays.net] has joined #wesnoth-dev
20100924 05:01:46-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 272 seconds]
20100924 05:03:39-!- Ivanovic_ is now known as Ivanovic
20100924 05:08:28-!- Blueblaze [~Blueblaze@adsl-99-171-161-94.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100924 05:08:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 05:30:57-!- Blueblaze [~Blueblaze@adsl-99-171-161-94.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100924 05:54:17-!- Blueblaze [~Blueblaze@adsl-99-171-161-94.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100924 06:10:15< billynux> Soliton, Ivanovic, around? any network related trouble after  Revision #46654 ?
20100924 06:22:04-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20100924 06:22:07-!- Gambit [~Gambit@unaffiliated/gambit] has quit [Remote host closed the connection]
20100924 06:30:05-!- Mythological [Ilccov@77.28.89.98] has quit [Quit: "goodbye"]
20100924 06:44:09-!- johndh [~johndh@adsl-85-59-158.mco.bellsouth.net] has joined #wesnoth-dev
20100924 07:10:15-!- Blueblaze [~Blueblaze@adsl-99-171-161-94.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100924 07:21:44-!- koan [~koan@unaffiliated/koan] has quit [Read error: Connection reset by peer]
20100924 07:21:53-!- koan [~koan@unaffiliated/koan] has joined #wesnoth-dev
20100924 07:22:06-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100924 07:37:37< Soliton> billynux: nothing so far but i don't see a reason why the pure virtual method call won't be back.
20100924 07:37:57< billynux> Soliton, Indeed
20100924 07:39:00< Soliton> you should check the lifetime of objects with that method. there should be a possibility for someone calling that function while the object is being destructed.
20100924 07:39:21< billynux> Soliton, that error could be related to the same thing in the following way: The method is a pure virtual one originally, but in no way the pointer used in could contain a base class
20100924 07:40:04< billynux> Soliton, yes, exactly. My guess is the object got destroyed, and then there was no way to call the handler method
20100924 07:40:16< billynux> debugging this, as you may imagine, is not so simple
20100924 07:40:57< Soliton> you have some UML overview of ANA, right? can you point me to where in there the problematic class/method is?
20100924 07:41:11< billynux> 1'
20100924 07:43:49< billynux> Soliton, http://imagebin.org/115458
20100924 07:46:08< billynux> well... wierd, asio_sender:127 is a call to handle_send, which I intended to be to the privately-declared handle_send
20100924 07:48:06-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20100924 07:49:51< billynux> Soliton, I wouldn't think the class diagram should help you much. You are dealing with a server (or client) instance that has been issued a send operation
20100924 07:50:17< billynux> what you say does make sense though (the object being destructed while the send operation completes)
20100924 07:50:39< Soliton> yeah, a sequence diagram would be better.
20100924 07:52:09< billynux> yeah, don't have one though :(
20100924 07:52:11< Soliton> so the server keeps pointers to the send_handler? who destroys the send_handler?
20100924 07:53:12< billynux> the send_handler is your (the user) responsibility
20100924 07:53:35< billynux> it should always be something that outlives the network operations
20100924 07:54:07< billynux> the server doesn't keep the pointer to it, the operation does (via the parameters set to the handlers)
20100924 07:55:00< billynux> just think that the network operation holds the pointer to the handler object while it is operating
20100924 07:55:04< Soliton> what's an operation?
20100924 07:55:28< billynux> sending somethingggggg
20100924 07:55:32< billynux> ah... sticky f
20100924 07:55:34< billynux> g!
20100924 07:55:41< billynux> :)
20100924 07:56:21< billynux> so... you have an ana::server* object, say server_, you do server_->send_all( ana::buffer( big_buffer ), &my_handler_obj );
20100924 07:57:00< billynux> my_handler_obj should inherit from ana::send_handler (see CD) and implement handle_send
20100924 07:57:23< billynux> asio_sender.cpp:127 is before that, it's the implementation code for all senders of ana
20100924 07:57:45< billynux> and it sends little batches of data while updating the statistics stuff
20100924 07:58:28< billynux> after it completes, it will call handler->handle_send(   <corresponding_error_code, etc > )
20100924 07:59:01< billynux> Soliton, ^
20100924 08:00:13< Soliton> so how long does the server have to keep my_handler_object? can you use a smart pointer for it?
20100924 08:00:35-!- johndh [~johndh@adsl-85-59-158.mco.bellsouth.net] has quit [Quit: Leaving]
20100924 08:01:45< billynux> you have to ensure that it lives long enough to the call to handle_send
20100924 08:01:59< billynux> if ANA is implemented right, this will always happen exactly once
20100924 08:02:33< billynux> but in this case, the error goes off in a call to a private method of asio_sender (not a call to the handler obj)
20100924 08:06:38< Soliton> "you have to ensure that it lives long enough to the call to handle_send" that call is async, right? you must have some idea on how the caller is supposed to accomplish that?
20100924 08:10:57< billynux> just that it will attempt to complete the operation and call you once after either completing, failing or timing out
20100924 08:13:24< billynux> Soliton, I know you complained the implementation wasn't really async. This is both true and false. ANA is async, but the current network API (that in network.hpp) is not. Thus, I implemented this synchronous API using an async lib
20100924 08:17:43-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]]
20100924 08:26:33-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: bbl]
20100924 08:33:12< Soliton> billynux: ok, but then there shouldn't be any issue with early destruction of the send_handler possible?
20100924 08:34:16< billynux> Soliton, there is always that issue. asio_sender:127 makes a call to a private function though
20100924 08:35:54-!- joo [~joo@unaffiliated/joo] has quit [Quit: College.]
20100924 08:41:54-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100924 08:41:54-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100924 08:41:54-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100924 08:49:19-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100924 08:49:49-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100924 08:49:56< billynux> YES! I got the connection through proxy thing working
20100924 08:53:43< Soliton> sweet, an easier way for trouble makers to get around bans on the server! :->
20100924 08:54:53< billynux> :D
20100924 08:56:52< CIA-35> billynux * r46681 /trunk/src/ (7 files in 2 dirs): (log message trimmed)
20100924 08:56:52< CIA-35> Added support to allow a client to connect through a proxy (using no
20100924 08:56:52< CIA-35> authentication or basic, see http://www.ietf.org/rfc/rfc2617.txt). Use it via
20100924 08:56:52< CIA-35> command line options --proxy, --proxy-address, --proxy-port, --proxy-user and
20100924 08:56:52< CIA-35> --proxy-password. All these options are undocumented yet and for testing
20100924 08:56:53< CIA-35> purposes at the moment, please feel free to document them (check the comments in
20100924 08:56:54< CIA-35> the new methods in network.hpp.) Example use: ./wesnoth --proxy-address
20100924 08:57:17< billynux> Soliton, mid-size commit ^. If you have a min check it out (I'm concerned about modifications to game.cpp, but everything appears to run just fine
20100924 08:58:08< billynux> I gotta go now, I'm fubar... and I have my most important job interview this Tuesday :S and the folks asked me for a mint-condition iPhone app I had to develop
20100924 08:58:20< billynux> sry, catharsis...
20100924 08:59:44< billynux> Ivanovic, see the proxy settings commit ^. And Soliton reports no recent server errors... although he is (rightfully) skeptical about quality of the rest of the code :)
20100924 09:03:43< Soliton> ana_send_handler handler;
20100924 09:03:43< Soliton> (*it)->client()->send( ana::buffer( base_char, size ), &handler);
20100924 09:03:43< Soliton> handler.wait_completion();
20100924 09:04:14< Soliton> billynux: ^ that is the part that makes send calls blocking and ensures the handler lives as long as the send call, right?
20100924 09:05:05< billynux> yes... not the best code I've produced. The send operations should have the timer values set, so it shouldn't block for good
20100924 09:05:50< billynux> but that is what I had to do to implement the sync API
20100924 09:07:23< Soliton> what makes you think that network::send_raw_data() used to be blocking?
20100924 09:07:50< billynux> it's supposed to return the amount of bytes sent
20100924 09:08:07< billynux> thinking out loud...
20100924 09:08:48< billynux> sorry, no, send_raw_data doesn't return anything
20100924 09:08:59< billynux> send_data does
20100924 09:09:54< billynux> hm...
20100924 09:09:59-!- happygrue [~George@wesnoth/developer/wintermute] has quit [Read error: Connection reset by peer]
20100924 09:10:15< Soliton> it's not blocking either.
20100924 09:10:27-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100924 09:10:42< billynux> ?
20100924 09:10:48< billynux> didn't parse
20100924 09:10:52-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100924 09:11:16< Soliton> send_data() is not blockig either.
20100924 09:11:27< Soliton> it returns the queued size.
20100924 09:11:50< billynux> Are you thinking about this: http://paste2.org/p/1001097 ?
20100924 09:12:49< Soliton> no idea. just seems there were some wrong assumptions made so far. :-)
20100924 09:14:30< billynux> maybe so... I tell you, changing this to a more async flavor will surely help code readability... can you play around with it? :)
20100924 09:14:44< Soliton> nope, i'm at work.
20100924 09:15:00< billynux> ok, I really can't do much until Tuesday
20100924 09:15:07< Soliton> you know the code though so i'm sure what you suggested makes sense.
20100924 09:15:35< billynux> what's the status of this network implementation so far? Are most clients using it? Are there any important servers running it?
20100924 09:15:44< Soliton> well, it's not urgent since apparently the server works so far. should be fixed before next stable though. ;-)
20100924 09:16:19< Soliton> trunk and dev server run it and all 1.9.0+ clients i assume.
20100924 09:16:49< Soliton> the load will not be much during the dev cycle even for the dev server probably.
20100924 09:17:09< Soliton> s/load/number of users/
20100924 09:17:59< Soliton> but i'm guessing the current implementation has issues once there will be some more users.
20100924 09:18:00< billynux> And whomever is running this is doing so with what failure rate approximately?
20100924 09:18:24< billynux> Soliton, I'm expecting them too. Concurrency is just ... hard
20100924 09:18:28< Soliton> failure rate == crashes per hour? :-)
20100924 09:18:37< billynux> any stats :P
20100924 09:18:43< billynux> stat
20100924 09:19:01< Soliton> on average it crashes once or twice a day, i think.
20100924 09:19:11-!- Ivanovic [~ivanovic@dtmd-4db2b956.pool.mediaWays.net] has quit [Changing host]
20100924 09:19:11-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100924 09:19:12< Soliton> but since one crash bug is fixed that should now be less.
20100924 09:19:50< Ivanovic> moin
20100924 09:19:55< billynux> hi Ivanovic
20100924 09:20:11< billynux> big chat with Soliton above that will surely interest you
20100924 09:21:17< billynux> Ivanovic, and... proxy support is here via (undocumented) command line options :) -> test away
20100924 09:21:24< billynux> I really gotta go guys
20100924 09:21:27< Soliton> for now most importantly the crash (i.e. pure virtual call) should get fixed. though if it's best to combine that with more async behaviour that's of course fine, too.
20100924 09:21:53< Soliton> billynux: ok, not to worry too much. ;-)
20100924 09:22:36< billynux> Soliton, to prevent these things I had the object destructors running the network threads wait for the thread termination
20100924 09:22:58< billynux> this should in principle ensure that no handler is called after that termination
20100924 09:23:04< billynux> Our case is a little different
20100924 09:23:13< billynux> since it involves only a call to a private method
20100924 09:23:24< billynux> anyway... I hope to be around... say Wednesday
20100924 09:23:27< Soliton> i'd think smart pointers is an easy way to ensure life time constraints.
20100924 09:23:38< billynux> and I agree
20100924 09:23:43-!- johndh [84aa09ae@gateway/web/freenode/ip.132.170.9.174] has joined #wesnoth-dev
20100924 09:23:43-!- Ivanovic changed the topic of #wesnoth-dev to: 1.8.5 planned for today (14th September) at ~18:00 GMT+2 | 159 bugs, 300 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100924 09:23:54< billynux> off I go, take care.... and... I do care, really :)
20100924 09:23:58-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100924 09:25:10-!- Upthorn [ogmar@adsl-75-26-198-81.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100924 09:26:29-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-153.telecom.by] has joined #wesnoth-dev
20100924 09:27:59-!- johndh [84aa09ae@gateway/web/freenode/ip.132.170.9.174] has quit [Ping timeout: 252 seconds]
20100924 09:28:43-!- Upth [~ogmar@adsl-75-26-198-81.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 252 seconds]
20100924 09:38:12-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100924 09:46:11< shadowmaster> mordante: http://forums.wesnoth.org/viewtopic.php?f=4&t=31725
20100924 09:46:35< shadowmaster> mordante: what do you think? My code only sets the maximum width and height of a font:ttext object
20100924 09:47:43< shadowmaster> mordante: and renders that directly (see src/storyscreen/render.c pp line 387 onwards in that method)
20100924 09:48:14< shadowmaster> I don't understand why it seems to wrap into a single line in some systems.
20100924 09:49:06< shadowmaster> maybe you could also ask the user in that thread to do something to check if it's a problem with ttext in general, or jus with my code?
20100924 09:51:01< shadowmaster> also, maybe you want your forum account to be renamed to match your IRC/gna username? ;)
20100924 09:51:55< shadowmaster> (this is totally not a bribe)
20100924 09:54:43-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 240 seconds]
20100924 09:55:41-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100924 10:20:20< shadowmaster> gods, there are still dozens of Feedback threads to change their ownership.
20100924 10:20:48< shadowmaster> I think I'll just make a SQL script for the  rest, also to normalize the thread titles :/
20100924 10:41:30-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100924 10:44:12-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: 4:43 am and I'm still here? what the-]
20100924 10:51:37-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 11:01:17-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: later]
20100924 11:04:37-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100924 11:08:35< CIA-35> espreon * r46682 /trunk/data/core/images/units/human-loyalists/ (8 files): Ran umcpropfix.
20100924 11:13:37< CIA-35> ivanovic * r46683 /trunk/po/wesnoth/vi.po: updated Vietnamese translation
20100924 11:13:40< CIA-35> ivanovic * r46684 /branches/1.8/po/wesnoth/vi.po: updated Vietnamese translation
20100924 11:25:38< fabi> shadowmaster: hi
20100924 11:27:06< fabi> shadowmaster: 3 questions: First, I have found a thread in which you recommended some addon campaigns to a user. I can't find that thread anymore. Can you point me to it or repeat the names?
20100924 11:28:23< fabi> shadowmaster: Second, are the reports to a user being deleted after some time again? If not I strongly suggest to do so. It's not a good idea to keep people marked with misbehavior that was long ago forever.
20100924 11:29:44< fabi> shadowmaster: Third, I have read that espreon wants to take a wesbreak for a quarter or half a year. Did he organize a substitute for his duties?
20100924 11:34:30< Soliton> user notes are dated, no? (this discussion should go to #wesnoth-dev-nonpublic)
20100924 11:48:48-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100924 11:50:17< fabi> Soliton: Sorry, I thought the matter would be innocuous enough for being discussed in public.
20100924 11:59:28-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100924 12:00:31-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100924 12:00:31-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100924 12:00:31-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100924 12:01:55-!- fabi is now known as fendrin
20100924 12:10:59-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has joined #wesnoth-dev
20100924 12:34:16-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100924 12:57:57-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100924 13:00:24-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100924 13:00:54< fendrin> esr: I still have problems with wmlscope. 
20100924 13:01:24< fendrin> esr: It's giving me warnings about "line 119: more than one definition of MULTIPLAYER_GOLD is visible here." 
20100924 13:01:43< fendrin> No idea where the duplicate definition is going to be.
20100924 13:02:10< esr> That's odd, I just ran wmlscope and didn't see that - thought you'd fixed it.
20100924 13:02:24< fendrin> esr: Give me some seconds to commit please.
20100924 13:04:08< CIA-35> fendrin * r46685 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/ (8 files): LoW: More gold macro multiplayer things not working arrggghhh
20100924 13:04:09< fendrin> esr: ^
20100924 13:04:29< esr> OK, will sync now and look at it.
20100924 13:04:39< esr> Syncing.
20100924 13:05:03< esr> Running wmlscope...
20100924 13:08:00< fendrin> esr: The engine should process the scenarios in lexical order, beginning with 01_The_Uprooting.cfg. So there can't be a second definition of MULTIPLAYER_GOLD around there. 
20100924 13:10:28< esr> fendrin: You've tripped over  a quirk in the way wmlscope work.  It builds a cross-reference table first and themn looks for multipley-defined symbols by looking at the reference structure, so definition order is irrelevwnt. 
20100924 13:10:49< esr> I have a directive for this case, I think.
20100924 13:10:55< esr> I'll find it.
20100924 13:13:33< esr> Aha!
20100924 13:15:02-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 264 seconds]
20100924 13:16:35< esr> fendrin: Problem solved.  You forgot a needed #undef and a definition leaked.
20100924 13:17:12< fendrin> esr: Is there a more verbose mode of wmlscope that told you where to search or did you see it by reviewing all scenarios?
20100924 13:17:39< CIA-35> esr * r46686 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/19_Costly_Revenge.cfg: Add an #undef at end of file to plug a definition leak.
20100924 13:17:46< esr> I grepped for MULTIPLAYER_GOLD
20100924 13:18:21< fendrin> hmmm, stupid me.
20100924 13:18:32< esr> Happens to all of us.
20100924 13:19:14< fendrin> Okay, thank you very much :-)
20100924 13:25:38-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100924 13:33:04< esr> fendrin.: there's a weirkdly garbaged line at the end of your Chapter One cfg.  It looks like this: "##u#n#def MULTIPLAYER_GOLD_04" and has no trailing linefeed.
20100924 13:33:35< fendrin> esr: Thank you, I will look after it.
20100924 13:33:47< esr> I'll fix it, but I think there is something strangely wronmg with your editing tools.
20100924 13:34:21< esr> Every time after you change LoW, wmllint trips over missing trailing newlines.
20100924 13:34:33< fendrin> esr: That was a workaround for wmlscope not recognizing ##undef for being commented out.
20100924 13:35:30< fendrin> esr: I use emacs with the wml mode. No idea how emacs can be told to add a newline to the end of file in every case.
20100924 13:36:19< esr> There's LISP configuration variable for that but I've forgotten the name of it.
20100924 13:40:23-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20100924 13:51:59< fendrin> esr: Isn't every configuration done to emacs to be found somewhere in .emacs or .emacs.d ?
20100924 13:52:21-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100924 13:52:34< esr> fendrin: Yes.
20100924 13:54:15< fendrin> esr: ;; (setq require-final-newline t) ?
20100924 13:54:26< fendrin> what is t?
20100924 13:55:13< fendrin> true?
20100924 13:55:21< esr> t is the LISP literal for logical true.  Like True in C99.
20100924 13:55:45< esr> And that is the correct variable.
20100924 13:58:54< fendrin> cool, I guess that won't happen again.
20100924 14:00:23-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100924 14:02:59-!- Gambit [~Gambit@unaffiliated/gambit] has joined #wesnoth-dev
20100924 14:12:21-!- johndh [84aa09ae@gateway/web/freenode/ip.132.170.9.174] has joined #wesnoth-dev
20100924 14:16:23-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has quit [Remote host closed the connection]
20100924 14:27:33-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-153.telecom.by] has quit [Remote host closed the connection]
20100924 14:47:02< CIA-35> zookeeper * r46687 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/10_Alone_at_Last.cfg: Balancing and other tweaks to 'Alone at Last'.
20100924 14:48:14< johndh> Good, that scenario is tough as hell. lol
20100924 14:52:34< CIA-35> zookeeper * r46688 /trunk/data/campaigns/Descent_Into_Darkness/maps/10_Alone_at_Last.map: Biggerized the player's starting castle by 1 hex.
20100924 14:52:51< zookeeper> yeah...if people still complain that it's too hard then i don't know what to do -.-
20100924 14:53:18-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100924 14:54:43< Rhonda> It's too hard.
20100924 14:56:32 * zookeeper slaps Rhonda with a baneblade
20100924 14:58:19< CIA-35> zookeeper * r46689 /branches/1.8/data/campaigns/Descent_Into_Darkness/ (maps/Alone_at_last.map scenarios/10_Alone_at_Last.cfg): Ported most of r46687 and r46688 to 1.8.
20100924 15:03:10-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100924 15:04:16-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20100924 15:22:28< CIA-35> zookeeper * r46690 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/10_Alone_at_Last.cfg: Made the paladins arrive on turn 8 and Dela on turn 11 on all difficulty levels.
20100924 15:22:59< CIA-35> zookeeper * r46691 /branches/1.8/data/campaigns/Descent_Into_Darkness/scenarios/10_Alone_at_Last.cfg: Ported r46690 to 1.8.
20100924 15:26:01-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Quit: Leaving]
20100924 15:37:19< CIA-35> zookeeper * r46692 /trunk/ (changelog players_changelog): Updated changelogs.
20100924 15:38:32< CIA-35> zookeeper * r46693 /branches/1.8/ (changelog players_changelog): Updated changelogs.
20100924 15:51:35-!- FAAB1 [~huajie@219.142.250.102] has joined #wesnoth-dev
20100924 15:53:16-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20100924 15:53:18-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 240 seconds]
20100924 15:59:27-!- eleazzaar [~jbjerk@ppp-70-226-224-2.dsl.spfdil.ameritech.net] has quit [Quit: eleazzaar]
20100924 16:00:18-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100924 16:08:37< Soliton> anyone using gcc 4.5+ and binutils 2.20+?
20100924 16:20:04-!- eleazzaar [~jbjerk@ppp-70-226-224-2.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20100924 16:23:08< CIA-35> zookeeper * r46694 /trunk/data/campaigns/Descent_Into_Darkness/scenarios/11_Descent_into_Darkness.cfg: Tweaked the cutscene slightly.
20100924 16:24:47< alink> hmmm "This web site at www.opengl.org has been reported as an attack site"  by Firefox / google  :-/
20100924 16:25:32< CIA-35> zookeeper * r46695 /trunk/data/campaigns/ (4 files in 2 dirs): Changed some unit.status.* values from on/off to yes/no.
20100924 16:26:03< alink> ^I bet it's from the DirectX guys ;)
20100924 16:31:46-!- _jbx_ [~jbailey@12.190.80.225] has joined #wesnoth-dev
20100924 16:37:25-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100924 16:42:48< Ivanovic> AI0867, alink, esr, fendrin, Gambit, ilor_, loonycyborg, Soliton, Upthorn, zookeeper, everyone else who cares: i plan to get 1.8.5 out in 75mins
20100924 16:43:05< Ivanovic> if you have anything left for the maybe last 1.8.x release, please get it commited asap
20100924 16:43:30 * Soliton got nothing.
20100924 16:43:40 * esr has nothing.
20100924 16:43:49< CIA-35> ivanovic * r46696 /branches/1.8/po/ (wesnoth/vi.po wesnoth-utbs/cs.po): updated Czech and Vietnamese translation
20100924 16:43:56< CIA-35> ivanovic * r46697 /trunk/ (6 files in 5 dirs): updated Czech and Vietnamese translation
20100924 16:44:46 * alink has decided that he has nothing (this 1.8 teleport glitch can live)
20100924 16:45:31< Ivanovic> :)
20100924 16:45:36< fendrin> Ivanovic: nothing left here.
20100924 16:46:21< Gambit> No problem here.
20100924 16:46:53< Soliton> how can my computer beep without the pcspkr module?!
20100924 16:48:28< Ivanovic> magick!
20100924 16:48:45< alink> snd_pcsp ?
20100924 16:49:21< alink> from http://www.thinkwiki.org/wiki/How_to_disable_the_pc_speaker_(beep!)
20100924 16:49:23< Soliton> don't have that either.
20100924 16:49:41< loonycyborg> Soliton: I have pc speaker disabled in kernel config and it never beeps for me.
20100924 16:50:00< loonycyborg> With the exception of POST beep naturally.
20100924 16:50:11< esr> BTW, everyone...go see this: http://xiph.org/video/index.shtml.en  "A Digital Video Primer for Geeks".  Brilliant.
20100924 16:51:38< Soliton> kernel config says "CONFIG_PCSPKR_PLATFORM=y; CONFIG_INPUT_PCSPKR=m; # CONFIG_SND_PCSP is not set"
20100924 16:51:45-!- stikonas_ is now known as stikonas
20100924 16:52:31< stikonas> esr: it is indeed good, I've seen it on planet.freedesktop.org
20100924 16:55:12< esr> I'm writing a rave review of it for my blog now.
20100924 17:04:00< zookeeper> Ivanovic, i got nothing more either.
20100924 17:11:23< esr> Uh oh.  svn up is hanfing.
20100924 17:11:27-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has joined #wesnoth-dev
20100924 17:11:33< esr> Hanging, I mean.
20100924 17:16:36-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 17:23:28-!- joo [~joo@188-223-201-81.zone14.bethere.co.uk] has joined #wesnoth-dev
20100924 17:23:28-!- joo [~joo@188-223-201-81.zone14.bethere.co.uk] has quit [Changing host]
20100924 17:23:28-!- joo [~joo@unaffiliated/joo] has joined #wesnoth-dev
20100924 17:46:49-!- anonymissimus [~chatzilla@HSI-KBW-095-208-007-247.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20100924 17:47:43-!- knotwork_ [~markm@142.177.178.49] has quit [Read error: Connection reset by peer]
20100924 17:49:17-!- FAAB1 [~huajie@219.142.250.102] has quit [Quit: Leaving.]
20100924 18:14:39-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100924 18:17:02-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100924 18:21:36< anonymissimus> I get compilation errors in CodeBlocks
20100924 18:22:57< alink> better paste the error (and mention your revision, if not current trunk)
20100924 18:23:22-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100924 18:25:56< anonymissimus> http://wesnoth.pastebin.com/XMHm500b
20100924 18:26:20< anonymissimus> msvc compiles that file fine, interesting :P
20100924 18:27:07< anonymissimus> current trunk
20100924 18:27:12< alink> wesbot: log 46681
20100924 18:27:14< wesbot> billynux * r46681 : Added support to allow a client to connect through a proxy (using no authentication or basic, see http://www.ietf.org/rfc/rfc2617.txt). Use it via command line options --proxy, --proxy-address, --proxy-port, --proxy-user and --proxy-password. All these options are undocumented yet and for testing purposes at the moment, please feel free to document them (check the comments in the new methods in network.hpp.) Example use: 
20100924 18:27:21< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=46681
20100924 18:27:42< alink> I bet on this one^
20100924 18:28:06< anonymissimus> sure
20100924 18:28:23< anonymissimus> this shows that it counts to use more than just one compiler
20100924 18:29:16< alink> well, if *everyone* (including users) used the same compiler, things would be simpler ;p
20100924 18:29:53< loonycyborg> anonymissimus: #include <stdexcept> will probably fix that.
20100924 18:30:08< alink> yeah
20100924 18:31:26< loonycyborg> Different compiler version will have different system headers include each other. Probably some other header included stdexcept for msvc.
20100924 18:37:30< anonymissimus> why does MSVC always throw so many warnings ? is this just because few devs use it ?
20100924 18:42:53< Soliton> you could also change it so a config::error is thrown like elsewhere.
20100924 18:48:56< CIA-35> anonymissimus * r46698 /trunk/src/network.cpp: Fixed built with CodeBlocks (compiler mingw32-g++-dw2.exe)
20100924 18:50:50< Soliton> and in the network code there is network::error.
20100924 18:51:55< eleazzaar> alink: is there any possibility we'll have to revert the color-shifted water?
20100924 18:52:33< eleazzaar> i.e. is it premature to delete the images and WML from the old method that's currently unused?
20100924 18:52:40< alink> eleazzaar: no, I don't think so, why ?
20100924 18:53:13< alink> well, with svn nothing is never really lost
20100924 18:54:57< alink> I think we will always support Image Path Functions, and thus having it for terrains will always make sense
20100924 18:55:38< eleazzaar> i was more concerned about performance
20100924 18:55:54< alink> and at least coloring terrain images will always be needed, so any change should keep that
20100924 18:56:10< anonymissimus> nothing is never = there is always at least one thing :P
20100924 18:56:53-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20100924 18:57:17< alink> eleazzaar: no worry there, as I said, the small extra cost (when present) can be optimized and is minor compared to other more common similar things
20100924 18:57:26< eleazzaar> alink: OK
20100924 18:58:08< eleazzaar> BTW i object to whatever happend to me terrain-graphic.cfgs
20100924 18:58:08< alink> * small extra memory cost and other gain already balance it enough
20100924 18:58:19< eleazzaar> it's been reformatted
20100924 18:58:34< eleazzaar> all double blank lines removed
20100924 18:58:42-!- Vorpal [~AnMaster@unaffiliated/anmaster] has quit [Read error: Connection reset by peer]
20100924 18:59:11< alink> wesbot: log 46664
20100924 18:59:12< wesbot> esr * r46664 : Reindent mainline.
20100924 18:59:12< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=46664
20100924 19:00:14< eleazzaar> i use whitespace to break up different sectcion
20100924 19:00:17< eleazzaar> sections
20100924 19:00:24-!- Vorpal [~AnMaster@unaffiliated/anmaster] has joined #wesnoth-dev
20100924 19:01:14< Soliton> curious that reindentation includes new line removal.
20100924 19:01:38< esr> eleazzaar: Sorry.  There are wmlindent directives you can use to guard sections of text.  
20100924 19:01:44< alink> maybe just revert it and ask esr to avoid this file, because it's a enough special case
20100924 19:02:12< eleazzaar> esr: what do i put to keep newlines from being removed?
20100924 19:02:18< eleazzaar> from the whole file?
20100924 19:03:11< Soliton> i don't see how that is a good idea anywhere.
20100924 19:03:52< esr> eleazzaar: I suppose you could wrap the whole thing in "#wmlindent: start ignoring" and "#wmlindent: start ignoring", butt it would be nice if you came up with a less shotgun approach.
20100924 19:03:59< eleazzaar> at least for me, strategically placed extra whitespace makes long and complicated files much easier to navigate
20100924 19:04:45< esr> I don't reindent for kicks.  I wrote that tool in response to a request to canonicalize whitespespace so reading WML would be less fatiguing.
20100924 19:05:26-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20100924 19:05:33< esr> Every exceotion you mark out partly defeats that purpose.
20100924 19:05:35< Soliton> new lines have nothing to do with indentation.
20100924 19:05:36< alink> mmh double white lines seems not fatiguing and are often useful
20100924 19:05:46< eleazzaar> well, in this case it is not achieving it's purpose
20100924 19:06:09< Soliton> it is misleading to have a tool named wmlindent mess with new lines.
20100924 19:06:17< esr> Soliton: Don't fixate on indentation, then, think of it as a whitespace canonicalizer.
20100924 19:06:33< Soliton> so you really think that is useful?
20100924 19:06:38< esr> And come up with a better name if you want.
20100924 19:06:45< eleazzaar> does man serve the indentation, or does the indentation serve man?
20100924 19:06:47< Soliton> i don't want to.
20100924 19:07:03< Soliton> i want to make the tool more useful and less annoying.
20100924 19:08:03< esr> I've been applying it fotr two and a half years without a peeop from anyone, so I'd say the annoyance factor is demonstrably low.
20100924 19:08:36< alink> eleazzaar: I also disagree with this change, but maybe you can use  "#"  or "#                      (section name)"
20100924 19:08:46< eleazzaar> i haven't been very active in the past 2.5 years
20100924 19:09:51< CIA-35> eleazar * r46699 /trunk/data/core/terrain-graphics.cfg: reverting the removal or useful newlines.
20100924 19:10:45< esr> Well, put in those directives or they'll get clobbered when I next reindent (usually about once per six months).
20100924 19:11:00< eleazzaar> so i put the same thing at start and end/
20100924 19:11:01< eleazzaar> ?
20100924 19:11:22< eleazzaar> esr: just want to make sure that is not a typo
20100924 19:11:34< eleazzaar> "irc://irc.freenode.net/#wmlindent: start ignoring" and "irc://irc.freenode.net/#wmlindent: start ignoring",
20100924 19:11:44< Soliton> esr: can you answer my question instead of focusing on the infallibility of your tool?
20100924 19:12:06< esr> What question?
20100924 19:13:09< Soliton> for your convenience: "so you really think that is useful?"
20100924 19:13:40< esr> Duh!  If I didn't think it was useful I would have removed it already.
20100924 19:13:50< Soliton> or to rephrase: what is useful about removing newlines?
20100924 19:14:43< alink> yeah I am not aware of any indent style controlling the number of blank lines
20100924 19:14:50< alink> http://en.wikipedia.org/wiki/Indent_style
20100924 19:15:04< esr> to jump over them.When I was wrting the tool I noticed that I found large gutters of whitespace visually distracting.  It was, like, an effort to 
20100924 19:15:19< esr> Sorry:
20100924 19:15:44< esr> When I was wrting the tool I noticed that I found large gutters of whitespace visually distracting. It was, like, an effort to jump over them 
20100924 19:16:25< eleazzaar> large chunks of white space can be a visual problem
20100924 19:16:41< eleazzaar> a couple lines usually are not-- especially when meaningfully placed
20100924 19:16:49< esr> eleazzaar: Sorry.  you want "#wmlindent: stop ignoring" at the end.
20100924 19:16:58< eleazzaar> esr: thansk
20100924 19:17:50< esr> eleazzaar: I agree, but I couldn't think of a rule other than "reduce to one line" that would be easy for users to mentally model.
20100924 19:18:19< eleazzaar> esr: so what do you mean by "large gutters"? 4+ lines?
20100924 19:18:44< esr> eleazzaar: Yes, certainly that much.
20100924 19:18:52< Soliton> good thing i'm not a WML writer. i would find the arbitrarily spread wml tools directives not only visually distracting.
20100924 19:19:16< alink> for example, players_changelog use 1 blank line between sections and 2 between version, and I think it's more readable
20100924 19:20:13-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 19:20:43< esr> alink: I agree that such conventions are useful for text that is not a highly structured markup.
20100924 19:21:23< eleazzaar> esr:  i think a less destructive approach would be to ignore newlines unless there are more than 3 or 4 in a row
20100924 19:21:33< alink> well if it's help readability for simple text, it probably helps for more complex stuff too ;)
20100924 19:22:03< esr> alink: But the markup is its own structure.
20100924 19:22:46< esr> eleazzaar: I see where you're gping, and I actually considewred doing something like that.
20100924 19:23:27< alink> and this structure is sometimes not very readable, blank lines is a sort of comment (a blank comment if you want) helping that
20100924 19:23:47< esr> But a magic number like 3 or 4 introduces arbitrariness and arguments that 0, 1 or infinity does not.
20100924 19:24:13< esr> So I went for 1.
20100924 19:25:02< alink> plus, it's another rule to agree and enforce. It is already hard to have a rule about tab (which at least had technical reasons)
20100924 19:25:22< esr> alink: Yes, exactly.
20100924 19:25:35< eleazzaar> then leave the newlines alone
20100924 19:25:51< alink> esr: I mean: don't add another rule if not really needed
20100924 19:26:11< alink> (a rule controlling blank lines)
20100924 19:27:07< esr> Without one, how do we eliminate distracting gutters?
20100924 19:28:33< alink> get used to it, it's not hard, and people will not add them without reasons, so should be rare and meaningful when present
20100924 19:28:41< zookeeper> wait, wmlindent collapses multiple linebreaks into one?
20100924 19:29:04< esr> zookeeper: Yes.  Has for the last... actually, three years.
20100924 19:29:54< esr> zookeeper is, BTW, the person who originally requested a mechanical indenter.
20100924 19:30:41< alink> indentation is not about blank line, the name is clear enough ;)
20100924 19:31:23< esr> Look at the function, not at the name.  The tool is justified by what it does, not what it's called.
20100924 19:32:13< Soliton> it doesn't seem like there is anymore justification than your preference.
20100924 19:32:31< zookeeper> i only recall asking for mainline to get indented in a uniform manner as a one-time procedure
20100924 19:32:32< alink> esr: ok I was just replying about the *indenter* request
20100924 19:33:01< zookeeper> besides, yes, i also pretty often use 2 linebreaks when i want to more clearly separate a piece of code from the rest
20100924 19:33:19< esr> Soliton: eleazzaar did note that gutters are distracting.
20100924 19:33:38< alink> useless big gutters
20100924 19:33:42< esr> So it's not just me.
20100924 19:34:01< anonymissimus> never seen in mainline
20100924 19:34:21< CIA-35> ivanovic * r46701 /trunk/po/wesnoth-httt/ja.po: updated Japanese translation
20100924 19:34:31< CIA-35> ivanovic * r46700 /branches/1.8/po/wesnoth-httt/ja.po: updated Japanese translation
20100924 19:34:32< esr> anonymissimus: That's because I nuked 'em!
20100924 19:37:26< anonymissimus> none would make that
20100924 19:37:26< zookeeper> esr, eleazzaar also notes that collapsing 2 linebreaks is annoying as well...
20100924 19:37:26< anonymissimus> and I agree with the most that multiple newlines are useful for organization of wml
20100924 19:37:26< alink> yes there is degree here and special cases, and since a tool can't guess the intention of the writer, better not touch it
20100924 19:37:26< zookeeper> so the fact that he finds 4+ linebreaks distracting is hardly justification for collapsing 2 linebreaks if he finds that too to be annoying :p
20100924 19:37:26< CIA-35> ivanovic * r46702 /branches/1.8/ (98 files in 25 dirs): 
20100924 19:37:26< CIA-35> pot-update (no new/changed strings, preparing for 1.8.5)
20100924 19:37:26< CIA-35> regenerated doc files
20100924 19:37:26< Ivanovic> everyone ready for 1.8.5?
20100924 19:37:26< esr> zookeeper: I chose the simplest possible rule, nit expecting that anyone would find any part of it annoying...and no one did fotr three years.
20100924 19:37:55< CIA-35> ivanovic * r46703 /branches/1.8/ (changelog configure.ac players_changelog src/wesconfig.h): bump version to 1.8.5
20100924 19:38:01< esr> OK.  If you don't like ye rule it has now, agree on a better one please.
20100924 19:39:31-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100924 19:39:34< alink> why make a rule about blanks lines ? No one does. My IDE has various auto indent style and custom options but I don't see any about multiple blank lines
20100924 19:39:35< anonymissimus> I doubt it's neccessary
20100924 19:41:17-!- Ivanovic changed the topic of #wesnoth-dev to: releasing 1.8.5, using r46703 | 159 bugs, 300 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100924 19:42:00< alink> and "no rule" is "the simplest possible rule" ;)
20100924 19:42:14< anonymissimus> zookeeper: I have the [terrain] tag ready and made it like you wanted
20100924 19:42:28< anonymissimus> I'd like to have silene a look at it
20100924 19:44:09-!- silene [~plouf@wesnoth/developer/silene] has quit [Ping timeout: 265 seconds]
20100924 19:44:25< fendrin> esr: Do you have a music collection?
20100924 19:44:25< zookeeper> esr, i don't see a need for a linebreak-affecting rule. or it can be behind some option so it can be run when lifting something to mainline; no one who works on mainline WML really screws up indentation badly, so running the tool on mainline usually just changes intentionally unconventional indentation.
20100924 19:45:31< esr> fendrin: Yes, why?
20100924 19:45:32< zookeeper> (although i don't use intentionally unconventional indentation precisely because i know someone will run wmlindent anyway at some point and change it)
20100924 19:45:41< zookeeper> (otherwise i would)
20100924 19:47:06< esr> Zctually, it's not true that nobody who works on mainline screws up indentation badly.  There used to be a mainline disaster area for that, and if you guessed UtBS you got it in one.
20100924 19:47:10< fendrin> esr: Ever came to the problem that music files from other parts of the world are named with strange filename encodings that don't get recognized by current linux distributions properly? My Amarok music player crashes when I order it to build the database. I am so sad without my music but can't figure out which file it is.
20100924 19:48:03< esr> fendrin: Havem't tripped overr that, no.  Word problem of that kind I've had is Thai character set on a Silly Fools album.
20100924 19:48:10< esr> s/Word/Worst/
20100924 19:48:24< esr> (The Silly Fools rock *hard*, BTW.)
20100924 19:48:25-!- knotwork [~markm@142.177.178.49] has joined #wesnoth-dev
20100924 19:48:34< Ivanovic> WTF?!?
20100924 19:48:46< Ivanovic> the xdelta from 1.8.4 to 1.8.5 has 50(!!!) MB
20100924 19:48:47< fendrin> Maybe strace would be a solution to find the bad file but I would like to have a tool that just converts them into utf8.
20100924 19:48:59< fendrin> Ivanovic: Can't be.
20100924 19:49:06< Ivanovic> fendrin: is
20100924 19:49:31 * esr is looking  at the wmlindent code
20100924 19:49:35< zookeeper> esr, i was talking of recent history, not the time before wmlindent was originally run on mainline
20100924 19:50:11< CIA-35> silene * r46704 /trunk/src/ (scripting/lua.cpp team.hpp): 
20100924 19:50:11< CIA-35> Reverted r46674 as it allows buffer underflow.
20100924 19:50:11< CIA-35> Properly fixed the original issue.
20100924 19:50:35< esr> It got a bit grotty after each of the three (3!) rewrite attempts :-)
20100924 19:50:48< Ivanovic> the difference in the source tarball size is 512M to 557M
20100924 19:51:16< fendrin> Ivanovic: Music or art backports?
20100924 19:51:27< Ivanovic> no backports from what i know
20100924 19:51:34-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has joined #wesnoth-dev
20100924 19:51:34-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has quit [Changing host]
20100924 19:51:34-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100924 19:51:35< Ivanovic> but many translation updates that should compress nicely
20100924 19:51:51< Ivanovic> argh, got the reason
20100924 19:51:57< Ivanovic> used my trunk folder for the export
20100924 19:52:07< zookeeper> eleazzaar, do you want to try to commit the lurker's wood bridge or should i have a go at it? if some of the images are >72x72 then it might need some (hopefully simple) extra magic.
20100924 19:52:46< silene> anonymissimus: which [terrain] code?
20100924 19:53:08< anonymissimus> the one for the [terrain] tag
20100924 19:53:18< anonymissimus> to lua + SLF
20100924 19:54:17< silene> was it committed? or should i look elsewhere?
20100924 19:54:49< fendrin> Ivanovic: Wasn't so wrong when I told you "can't be".
20100924 19:55:08< Ivanovic> okay, the real only only has 2.1MB
20100924 19:55:12< Ivanovic> sounds more reasonable...
20100924 19:55:21< anonymissimus> http://wesnoth.pastebin.com/yJrse8cP part 1
20100924 19:57:30< Ivanovic> at least my system is powerful enough to compile the release stuff in <5min
20100924 19:57:52< fendrin> Ivanovic: 4 cores?
20100924 19:57:53< Ivanovic> the part that takes most time is uploading
20100924 19:57:57< Ivanovic> Sysinfo for 'rechner1': Linux 2.6.35.4 running KDE Development Platform 4.5.1 (KDE 4.5.1), CPU: AMD Phenom II X6 1055T at 2800 MHz (5625 bogomips), HD: 319/1250GB, RAM: 2868/3960MB, 228 proc's, 2.11h up
20100924 19:58:00< Ivanovic> six by now
20100924 19:58:02< Ivanovic> ;)
20100924 19:58:39< fendrin> My machine can do it in under <5 mins with 4 cores only but that is because I replaced the harddisk with a ssd.
20100924 19:59:02< anonymissimus> http://wesnoth.pastebin.com/YzCU6PVu part 2
20100924 19:59:12< Ivanovic> no ssd, plain normal harddrives
20100924 19:59:19< fendrin> With my harddisk the cores hadn't been on full load. Only ~80%
20100924 19:59:27< anonymissimus> I hardly used wesnoth.set_terrain due to the missing overlay option
20100924 19:59:30< silene> anonymissimus: the c++ part seems fine, though i have no clue what replace_if_failed is and if it's worth exporting; you may want to change the type of the "layer" variable and the == comparisons to strcmp(...) == 0 to avoid creating useless dynamic strings
20100924 19:59:37< Ivanovic> compiling a kernel (x86_64) with defconfig and no ccache active takes <2mins
20100924 20:00:19< anonymissimus> I don't quite understand replace_if_failed to but it's needed for [terrain]
20100924 20:01:24< silene> anonymissimus: in the lua code, replace literal by shallow_literal, in case someone puts a complex slf (no need to recursively copy it)
20100924 20:02:46< silene> so no visible issue, expect for these two optimizations
20100924 20:03:10< CIA-35> ivanovic * r46705 /tags/1.8.5/: tag 1.8.5 (using r46703)
20100924 20:03:14< anonymissimus> I argued with zookeeper that we should rename terrain=
20100924 20:03:38< anonymissimus> because so we need an [and] tag to filter for terrain
20100924 20:03:59< silene> another optimization: no need to check for the 5th parameter before taking its boolean since it defaults to false
20100924 20:04:12< silene> anonymissimus: it would break backward compatibility, wouldn't it?
20100924 20:04:18< anonymissimus> yes
20100924 20:04:45< anonymissimus> well, some deprecation message, wmllint fix...
20100924 20:06:16< zookeeper> anonymissimus, so does your patch work according to my idea or differently?
20100924 20:06:29< anonymissimus> according to you now
20100924 20:06:37< zookeeper> all right. fine by me ;)
20100924 20:07:08< zookeeper> eleazzaar, lurker's diagonal wood bridges really look very narrow, especially compared to the vertical ones
20100924 20:07:21< CIA-35> ivanovic * r46706 /branches/1.8/ (changelog configure.ac players_changelog src/wesconfig.h): post release version bump to 1.8.5+svn
20100924 20:07:35< Ivanovic> crimson_penguin, Rhonda, loonycyborg: new stable release done, currently uploading the sources
20100924 20:07:45< crimson_penguin> oh!
20100924 20:08:17-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100924 20:08:47-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100924 20:11:11-!- stikonas_ [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100924 20:11:11-!- stikonas_ [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100924 20:11:11-!- stikonas_ [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100924 20:12:26-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 265 seconds]
20100924 20:13:25< eleazzaar> zookeeper: if you want to commit the bridges please feel free.  I have plenty i need to finish up
20100924 20:13:46-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]]
20100924 20:13:47< eleazzaar> this was was designed to fit the old WML so it should all be 72x72
20100924 20:14:21< zookeeper> yeah, so it seems
20100924 20:14:27< eleazzaar> zookeeper: the narrowness bothers me some, but they are still better than the old ones
20100924 20:14:48< eleazzaar> it might not look so bad in another context
20100924 20:15:08< eleazzaar> where the prespective was more obvious
20100924 20:15:37-!- wesbot changed the topic of #wesnoth-dev to: releasing 1.8.5, using r46703 | 160 bugs, 300 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100924 20:17:24< CIA-35> silene * r46707 /trunk/data/lua/wml-tags.lua: Avoided string concatenation.
20100924 20:17:27< CIA-35> silene * r46708 /trunk/src/ (team.cpp team.hpp): Exported controller strings.
20100924 20:17:29< CIA-35> silene * r46709 /trunk/src/scripting/lua.cpp: Added Lua accessors for side controllers.
20100924 20:17:33< CIA-35> silene * r46710 /trunk/data/lua/wml-tags.lua: Simplified code a bit.
20100924 20:20:22-!- Ivanovic changed the topic of #wesnoth-dev to: released 1.8.5, announcing "soon" | 160 bugs, 300 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100924 20:23:08< CIA-35> zookeeper * r46711 /trunk/data/core/images/terrain/bridge/ (23 files): Added lurker's new wooden bridges.
20100924 20:26:40< CIA-35> anonymissimus * r46712 /trunk/src/scripting/lua.cpp: Epanded wesnoth.set_terrain to accept optional 4th (layer) and 5th (replace_if_failed) arguments. Allows for moving [terrain] to lua.
20100924 20:26:45< eleazzaar> zookeeper: thanks
20100924 20:27:35< zookeeper> shadowmaster, http://www.wesnoth.org/forum/viewtopic.php?p=457895#p457895 <- can you up lurker's attachment size limit or something?
20100924 20:29:57< CIA-35> anonymissimus * r46713 /trunk/ (data/lua/wml-tags.lua src/game_events.cpp): 
20100924 20:29:57< CIA-35> Moved [terrain] to lua. Added support for SLF. Note that the SLF's terrain= key
20100924 20:29:57< CIA-35> is not valid; terrain= specifies the new terrain instead like it used to do.
20100924 20:29:57< CIA-35> Filtering for terrain can be done with the [terrain][and]terrain= workaround.
20100924 20:32:13-!- alink [~alink@wesnoth/developer/alink] has quit [Ping timeout: 245 seconds]
20100924 20:32:48< CIA-35> anonymissimus * r46714 /trunk/changelog: changelog update for previous commit
20100924 20:35:13< CIA-35> silene * r46715 /trunk/src/scripting/lua.cpp: Added a 'valid' field to units.
20100924 20:35:16< CIA-35> silene * r46716 /trunk/data/lua/wml-tags.lua: Simplified code a bit.
20100924 20:38:24< CIA-35> eleazar * r46717 /trunk/data/core/ (209 files in 2 dirs): cleaning up after the auto-colored water change.
20100924 20:39:22< anonymissimus> silene: buffer underflow: the index could have become -1 ?
20100924 20:39:45< silene> anonymissimus: yes, 0 - 1
20100924 20:40:04< anonymissimus> how can it be FAAB and me overlooked that :(
20100924 20:40:56< silene> yes, that surprised me a bit, especially since the fix could have been reduced to just adding "+1" (which i eventually did)
20100924 20:41:30-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20100924 20:41:45-!- alink [~alink@109.88.8.143] has joined #wesnoth-dev
20100924 20:41:45-!- alink [~alink@109.88.8.143] has quit [Changing host]
20100924 20:41:45-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100924 20:42:45-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100924 20:43:34-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100924 20:45:16< CIA-35> alink * r46718 /trunk/src/ (23 files in 6 dirs): Move all calls of SDL_BlitSurface into one place (sdl_blit)
20100924 20:45:33< CIA-35> alink * r46719 /trunk/src/ (14 files in 4 dirs): Move all calls to SDL_FillRect into one place (sdl_fill_rect)
20100924 20:45:46-!- johndh [84aa09ae@gateway/web/freenode/ip.132.170.9.174] has quit [Quit: Page closed]
20100924 20:48:48-!- stikonas_ is now known as stikonas
20100924 20:49:23< anonymissimus> but unsigned = something - 1 is save ?!
20100924 20:49:59< CIA-35> eleazar * r46720 /trunk/data/core/terrain-graphics.cfg: additional cleaning up after the change to auto-colored water.
20100924 20:54:52 * Rhonda steps onto zookeeper's toes
20100924 20:55:05< silene> anonymissimus: yes, it's modulo arithmetic, so it wraps to the biggest positive (!) unsigned integer
20100924 20:55:25< fendrin> eleazzaar: Can you do a hill terrain that is between the old and the new one color wise? Even if there would be a nicer transition between them, it's hard to to a good looking map with all that different dirt and grass now.
20100924 20:59:31< eleazzaar> fendrin: can you explain a little?
20100924 21:00:42-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100924 21:02:15< fendrin> eleazzaar: Okay, I will try. The both hill terrain types don't look good to each other, the color change is too abrupt. Maybe that can be fixed by giving them a better transition to each other but still there is much more color possible for flat terrain now than for hills. So I guess the best solution would be to have a hill terrain that corresponds to a grass terrain type.
20100924 21:02:22-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20100924 21:02:22-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20100924 21:02:22-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100924 21:02:30< fendrin> hi mordante
20100924 21:02:33< mordante> servus
20100924 21:02:56< mordante> fendrin, you were looking for me
20100924 21:03:34< fendrin> mordante: Yes, I have a feature request that is mostly consists of work on the lobby gui.
20100924 21:04:09< mordante> fendrin, and what is the request?
20100924 21:04:25< fendrin> mordante: https://gna.org/bugs/?16748
20100924 21:05:05< zookeeper> Rhonda, huh?
20100924 21:05:09< fendrin> mordante: I think that I can handle all wml related stuff myself, it's only the gui part that bothers me.
20100924 21:05:26< Ivanovic> hmm, what are the four most common hotkeys by "normal" players during normal gameplay?
20100924 21:05:42< fendrin> mordante: I mean the processing of the wml in c++ and the loading of the wml tree.
20100924 21:05:46< Ivanovic> are those: next unit, end turn, recruit, recall ?
20100924 21:05:52< Rhonda> zookeeper: Because you slapped me!
20100924 21:06:03< alink> Ivanovic: undo ?
20100924 21:06:23-!- kitty_ [~kathrinpo@e180198217.adsl.alicedsl.de] has joined #wesnoth-dev
20100924 21:06:28< fendrin> hi kitty_
20100924 21:06:29< alink> Ivanovic: and recall is not used in MP
20100924 21:06:35< eleazzaar> Ivanovic: those 4 are probably my main ones
20100924 21:06:38< kitty_> hi!
20100924 21:06:48< Mica> Hi kitty_
20100924 21:06:50< mordante> hi kitty_ 
20100924 21:07:11< Ivanovic> hmm, i could put next turn to space, though this is problematic regarding accidental clicks, too...
20100924 21:07:32< mordante> fendrin, I've to see, my plate is rather full at the moment and time somewhat limited
20100924 21:08:19< mordante> Ivanovic, nothing for 1.8.5 for me
20100924 21:08:21< fendrin> mordante: Maybe we can work together to fasten things up.
20100924 21:09:27< Ivanovic> mordante: would be too late anyway
20100924 21:09:29< Ivanovic> ;)
20100924 21:09:58< CIA-35> mordante * r46721 /trunk/src/network.cpp: Remove unused variables.
20100924 21:10:04< fendrin> eleazzaar: I mean a corresponding hill terrain to each grass terrain type.
20100924 21:10:18< mordante> anonymissimus, ^^
20100924 21:10:18-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
20100924 21:10:34< eleazzaar> yeah, i understand what you mean
20100924 21:10:36< anonymissimus> ?
20100924 21:10:48< mordante> anonymissimus, fixes your compilation issue
20100924 21:11:02< anonymissimus> ok
20100924 21:11:28< mordante> fendrin, not sure, but we can see
20100924 21:11:28< eleazzaar> fendrin: there's probably room for another hill color
20100924 21:11:34< mordante> Ivanovic, ok
20100924 21:11:52< eleazzaar> but i don't want to match the grass colors:  dry-grass colored hills would look too much like desert
20100924 21:12:09< fendrin> mordante: I will sent a mail to the dev-ml. To get a good definition of what features should be supported.
20100924 21:12:22< mordante> ok
20100924 21:13:12< mordante> anonymissimus, nevermind looked at the wrong thing, my commit is unrelated
20100924 21:13:20< anonymissimus> lol
20100924 21:13:29< anonymissimus> hehe
20100924 21:13:35< anonymissimus> ;)
20100924 21:19:20< CIA-35> alink * r46722 /trunk/src/attack_prediction_display.cpp: 
20100924 21:19:20< CIA-35> Kill the only use of SDL Color Key (in damage calculations)
20100924 21:19:20< CIA-35> Also use a non-transparent (and slightly darker) background behind colored bars
20100924 21:19:20< CIA-35> because I think it looks better and preserving old look was more work.
20100924 21:19:38< anonymissimus> um how do I document replcace_if_failed when I don't understand it :(
20100924 21:19:56< anonymissimus> is it fine to write "see the tag docs" :P
20100924 21:30:16-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Read error: Connection reset by peer]
20100924 21:37:59< anonymissimus> zookeeper: may I ask you...can you think of tags left that need a SLF ?
20100924 21:39:41< fendrin> anonymissimus: [remove_overlay_item}
20100924 21:39:58< fendrin> anonymissimus: Sorry, that needs a SUF.
20100924 21:40:18< anonymissimus> um I don't even know that tag ;)
20100924 21:41:29< anonymissimus> are you sure that exists ? or do you mean remove_unit_overlay ? that has a SUF
20100924 21:41:41< anonymissimus> and unit_overlay too
20100924 21:42:01< fendrin> anonymissimus: Yes and Yes. Someone must have done it during the last week.
20100924 21:42:48-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100924 21:42:50< zookeeper> umm...
20100924 21:42:51< anonymissimus> I won't do [item], for that one it doesn't make much sense
20100924 21:42:58< zookeeper> oh, right, nevermind.
20100924 21:43:11< zookeeper> anonymissimus, i'll have a look
20100924 21:44:39-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100924 21:44:51< zookeeper> anonymissimus, nope, can't see any others needing SUF
20100924 21:44:58< zookeeper> oh, wait, you wanted SLF
20100924 21:45:07 * zookeeper slaps himself
20100924 21:45:08< anonymissimus> both
20100924 21:45:36 * Rhonda slaps zookeeper, too
20100924 21:45:39< anonymissimus> these tags are a pain, I look in the refernece wml after testing that it doesn't work to see that the filter isn't supported
20100924 21:45:48 * Mica slaps zookeeper for the hell of it
20100924 21:47:17< loonycyborg> Ivanovic: 1.8.5 windows installer is uploaded.
20100924 21:47:49< Ivanovic> loonycyborg: cool
20100924 21:48:03< zookeeper> anonymissimus, [store_villages] is the only one i can think of, but i'm not sure if we even want to keep that tag since [store_locations] can do its job just fine...at least as long as all villages follow the *^V* terrain string convention.
20100924 21:52:59< anonymissimus> ok so I won't do store_villages probably, although it should be relatively easy to do in lua
20100924 21:53:35< fendrin> anonymissimus, zookeeper: What about [unit] ?
20100924 21:53:53-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has left #wesnoth-dev []
20100924 21:53:59< anonymissimus> [unit] with a SLF ??
20100924 21:54:09< fendrin> Right.
20100924 21:54:13< anonymissimus> hm
20100924 21:54:21 * zookeeper shrugs
20100924 21:54:46< anonymissimus> you have usecases for that ?
20100924 21:55:07< fendrin> Tomato surprise ambushes for free.
20100924 21:55:45< zookeeper> fendrin, you mean that it'd place a unit on every matching location? or that it'd place one unit on one of the matching locations picked at random?
20100924 21:56:04< zookeeper> the latter could be confusing and the former would be crazy ;)
20100924 21:56:05< fendrin> zookeeper: Right, maybe both depending on an extra attribute.
20100924 21:56:19< anonymissimus> on every matching is somehow bad, you'd get id conflicts
20100924 21:57:04< fendrin> anonymissimus: You don't need to specify id. The engine will give out generic ones.
20100924 21:57:09-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 21:57:30< anonymissimus> in case you would specify an id
20100924 21:57:58< anonymissimus> hm well I think that's rather a not-so frequent usecase
20100924 21:58:48< anonymissimus> you should write some own wml code for that I think
20100924 21:59:11< fendrin> A usecase could be to spawn a unit far enough away from all player units to not be in reach in one turn for example.
20100924 21:59:29< fendrin> Or spawning it in a region that is covered by shroud/fog.
20100924 21:59:35< fendrin> And and and and and much more.
20100924 22:00:49< anonymissimus> well during all my unit creations so far I didn't wish form a SLF in [unit]
20100924 22:01:09< anonymissimus> the code of [unit] looks too complicated for me, won't do that sry
20100924 22:01:33< zookeeper> fendrin, well if you're doing something that complicated then it doesn't add all that much lines to store the locations separately, pick a random index and then use that...besides you'd need an extra [if] with the whole SLF anyway to catch the possibility of there being no such location
20100924 22:01:51< zookeeper> better to just store the locations, check if any were stored and then place the unit on a random one
20100924 22:02:09< fendrin> Well, that is the case for all missing slf, isn't it?
20100924 22:02:44< fendrin> With that argument we can remove most slf from most tags and just handcode it.
20100924 22:03:39< anonymissimus> but for the tags I did it we agree that it's needed
20100924 22:04:30< fendrin> Well, if you don't see the need for unit... I could use it in at least 3 of the scenarios I maintain.
20100924 22:04:42< anonymissimus> besides, units are somehow "individuals", it doesn't feel right to place severel at once
20100924 22:04:57< anonymissimus> this automatic adding of an if etc
20100924 22:05:03< anonymissimus> id
20100924 22:05:20< zookeeper> esr, btw, i told kitty that we'd like to turn relnan (in liberty) female, i presume you're still up for that.
20100924 22:05:26< fendrin> anonymissimus: Placing one on the first matching location or at a random one is enough.
20100924 22:05:42< esr> zookeeper: Yes.
20100924 22:06:14< anonymissimus> well I would probably write some lua code for it if I needed that functionality
20100924 22:06:46< fendrin> anonymissimus: And the units being individuals is still granted, they can all get their own id, random name, random gender and random traits.
20100924 22:07:42< fendrin> anonymissimus: [sound_source] is still missing a SLF
20100924 22:08:30< anonymissimus> I thought about that too, but sound sources should also stay x,y similar to [item]
20100924 22:08:54< anonymissimus> several sound sources nearvy don't make much sense
20100924 22:09:01< anonymissimus> nearby
20100924 22:09:47< fendrin> anonymissimus: Why? One can filter for fireplace for example and have every hex containing one it's own source.
20100924 22:10:32< silene> i thought fireplace was an overlay
20100924 22:11:22< fendrin> Yes, but most campaigns that use it regularly define a own terrain for it. Like UtbS for example.
20100924 22:11:55< fendrin> That would lead to a new feature request, SLF should be able to filter for overlays. That would be really come in handy.
20100924 22:14:24-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100924 22:15:09< fendrin> AI0867: Did you commit the [tunnel] feature to trunk?
20100924 22:15:33< fendrin> AI0867: I have not done any ai performance test on it until now.
20100924 22:23:33< anonymissimus> SLF can filter for overlays; terrain=*^xy
20100924 22:25:15< fendrin> anonymissimus: I meant the overlay like silene used the word. It is placed with [item].
20100924 22:35:28-!- johndh [~johndh@adsl-85-59-158.mco.bellsouth.net] has joined #wesnoth-dev
20100924 22:41:35-!- kitty_ [~kathrinpo@e180198217.adsl.alicedsl.de] has quit [Quit: good night all]
20100924 22:44:18-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 22:44:46-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100924 22:45:47< anonymissimus> fendrin: http://wesnoth.pastebin.com/PLiJBXZG that adds SLF to [unit]
20100924 22:46:30< anonymissimus> is uses the existing [unit], quick and dirty :P
20100924 22:47:40-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 272 seconds]
20100924 22:47:40-!- noy_ is now known as noy
20100924 22:53:16-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20100924 22:54:35< fendrin> anonymissimus: nice
20100924 22:55:18< anonymissimus> that way it's easy, but not acceptible for wml_tags.lua I guess
20100924 22:59:18-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100924 23:00:08-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 245 seconds]
20100924 23:04:38< fendrin> anonymissimus: Hmmm, why not? It seems compatible to the old one. Maybe a check for the presence of the filter or the semi filter is needed.
20100924 23:07:17< anonymissimus> fendrin: hm it splits up the implementation of [unit] into C++ and lua, thats a bit bad
20100924 23:08:03< anonymissimus> and silene doesn't add such modify-existing-event-handlers implementations so I guess simething is bad about it
20100924 23:08:03< fendrin> anonymissimus: That is the case for [objectives] and [message] as well, or has this been changed?
20100924 23:08:18-!- noy_ [~Noy@S010600035203c8ac.vw.shawcable.net] has joined #wesnoth-dev
20100924 23:08:18-!- noy_ [~Noy@S010600035203c8ac.vw.shawcable.net] has quit [Changing host]
20100924 23:08:18-!- noy_ [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100924 23:09:57< zookeeper> really if you want that functionality then i'd rather just have a pick_one_at_random=yes key [store_locations]
20100924 23:10:21< anonymissimus> message is a special case I guess; e.g. it's not possible in lua to do the ESC-key leave out messages -thing
20100924 23:10:41< fendrin> I can live on without that that feature. 
20100924 23:11:06< fendrin> But SLF being able to filter for [item] would be useful.
20100924 23:11:37< fendrin> And I can't understand why [item] and [soundsource] shall not support full SLF themselves.
20100924 23:11:49< anonymissimus> then add them
20100924 23:12:09< fendrin> I don't have much luck with getting my c++ into trunk.
20100924 23:12:17< anonymissimus> the trick with existing event handlers should work for both :P
20100924 23:12:22-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 272 seconds]
20100924 23:12:22-!- noy_ is now known as noy
20100924 23:12:28< fendrin> And I don't have the time to waste in code that is not going to be used.
20100924 23:13:20< anonymissimus> gm although item and sound_source can appear at the scenario toplevel, thats a problem
20100924 23:14:00< fendrin> anonymissimus: Yes, maybe at that time the engine is not ready to do SLFs.
20100924 23:14:17< anonymissimus> it does for time_area so it is
20100924 23:17:33< mordante> I'm off night
20100924 23:17:43-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100924 23:21:45-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Read error: Operation timed out]
20100924 23:22:47-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has joined #wesnoth-dev
20100924 23:25:07-!- Blueblaze [~Blueblaze@99.171.161.94] has joined #wesnoth-dev
20100924 23:25:32< CIA-35> zookeeper * r46723 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/01_The_Uprooting.cfg: Fixed an #ifdef MEDIUM to #ifdef NORMAL.
20100924 23:28:25< anonymissimus> the code for sound_source is very different than the one for the other action handlers
20100924 23:28:59< anonymissimus> for item is seems doable but I don't understand that lua part well emough
20100924 23:29:07< fendrin> zookeeper: Thank you :-)
20100924 23:36:29< zookeeper> ohh, there's a new 1.8 campaign which isn't entirely silly.
20100924 23:36:35 * zookeeper plays
20100924 23:42:21-!- alink [~alink@wesnoth/developer/alink] has quit [Read error: Operation timed out]
20100924 23:42:35-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20100924 23:43:23-!- koan [~koan@unaffiliated/koan] has quit [Read error: Connection reset by peer]
20100924 23:43:27-!- koan_ [~koan@unaffiliated/koan] has joined #wesnoth-dev
20100924 23:48:14< anonymissimus> could someone please join 1.9 multiplayer server ? wnt to know whether my built is wring or something
20100924 23:48:29< anonymissimus> it's always empty when I'm there
20100924 23:54:59-!- koan_ [~koan@unaffiliated/koan] has quit [Remote host closed the connection]
20100924 23:56:56-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100924 23:59:29-!- _jbx_ [~jbailey@12.190.80.225] has quit [Quit: Ha ha, charade you are.]
--- Log closed Sat Sep 25 00:00:31 2010