--- Log opened Sun Oct 03 00:00:25 2010
20101003 00:06:07-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 240 seconds]
20101003 00:26:50< Ivanovic> mordante: it happens every now and then that it is broken because some langs do send in borked manpages or manuals
20101003 00:28:03< Ivanovic> mordante: and it seems to be working over here using r46872
20101003 00:28:10< Ivanovic> (though something in the indonesian manual is borked)
20101003 00:30:48< zookeeper> http://www.wesnoth.org/forum/viewtopic.php?p=459417#p459417 <- am i being silly for making such a stab in the dark, or couldn't that be done simply by adding a new case to construct_dialog.cpp:120-145 and then using that in manager.cpp:1260?
20101003 00:31:03< zookeeper> just curious.
20101003 00:38:44< shadowmaster> no idea.
20101003 00:39:19< shadowmaster> Mist also said that porting the Wesnoth navbar to prosilver would be a time-consuming task back when he was the forum admin, so take that as you will.
20101003 00:56:30-!- nejucomo [~Adium@70-36-197-40.dsl.dynamic.sonic.net] has quit [Quit: Leaving.]
20101003 01:03:21-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20101003 01:04:54-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20101003 01:14:20-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20101003 01:23:40-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101003 01:23:42-!- boucman [~rosen@wesnoth/developer/boucman] has left #wesnoth-dev []
20101003 01:28:48-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20101003 01:50:09-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20101003 01:56:16-!- Blueblaze [~Blueblaze@99.148.247.46] has joined #wesnoth-dev
20101003 02:15:37-!- wesbot changed the topic of #wesnoth-dev to: 157 bugs, 298 feature requests, 14 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20101003 02:28:04-!- nejucomo [~Adium@70-36-197-40.dsl.dynamic.sonic.net] has joined #wesnoth-dev
20101003 02:33:04-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101003 02:38:04-!- Octalot [~noct@host86-166-161-16.range86-166.btcentralplus.com] has quit [Ping timeout: 252 seconds]
20101003 02:57:50-!- nejucomo [~Adium@70-36-197-40.dsl.dynamic.sonic.net] has quit [Quit: Leaving.]
20101003 02:58:40-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 03:21:47-!- Upth [ogmar@adsl-75-26-162-59.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20101003 03:24:48-!- Upthorn [ogmar@adsl-75-26-194-80.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20101003 03:47:30-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has joined #wesnoth-dev
20101003 03:53:03-!- Vetinari [~lukjad@unaffiliated/lukjad] has quit [Ping timeout: 240 seconds]
20101003 04:05:57-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20101003 04:25:27-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20101003 04:27:27-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20101003 04:34:35-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 04:45:53-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101003 04:47:00-!- Ivanovic_ [~ivanovic@dtmd-4db2d1ba.pool.mediaWays.net] has joined #wesnoth-dev
20101003 04:49:38-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 272 seconds]
20101003 04:50:56-!- Ivanovic_ is now known as Ivanovic
20101003 04:55:38-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20101003 05:08:55-!- hhyloc [~hhyloc@113.166.188.249] has joined #wesnoth-dev
20101003 05:27:50-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20101003 05:27:52-!- Gambit [~Gambit@unaffiliated/gambit] has quit [Remote host closed the connection]
20101003 05:30:34-!- yenloc [~hhyloc@113.166.188.249] has joined #wesnoth-dev
20101003 05:33:46-!- hhyloc [~hhyloc@113.166.188.249] has quit [Ping timeout: 240 seconds]
20101003 05:55:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 06:00:06-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 240 seconds]
20101003 06:13:23-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has quit [Quit: And that's the end of THAT chapter]
20101003 06:19:06-!- noy [~Noy@70.70.255.54] has joined #wesnoth-dev
20101003 06:19:07-!- noy [~Noy@70.70.255.54] has quit [Changing host]
20101003 06:19:07-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 06:23:28-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 245 seconds]
20101003 06:33:34-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has joined #wesnoth-dev
20101003 06:33:34-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has quit [Changing host]
20101003 06:33:34-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20101003 06:56:49< CIA-35> silene * r46885 /trunk/src/addon/manager.cpp: Changed the addon dialog buttons to 'Download' and 'Done'.
20101003 06:57:43< silene> zookeeper: yes, it's as simple as what you suggest, but even your suggestion is still way too complicated; gui1 was flexible enough so that modifying the internals is not even needed, see ^
20101003 07:04:28-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 264 seconds]
20101003 07:14:45-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20101003 07:16:25-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 07:18:53-!- shadowm_laptop2 [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101003 07:19:20-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has quit [Disconnected by services]
20101003 07:19:46-!- shadowmaster [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101003 07:21:23-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 276 seconds]
20101003 07:22:25< shadowmaster> silene: awesome, thanks.
20101003 07:23:11< silene> it's unfortunate this wording issue wasn't brought up earlier, it would have been fixed much sooner
20101003 07:23:58< shadowmaster> nah, it's just my fault for treating GUI1 as a blackbox.
20101003 07:24:37< silene> shadowmaster: note that i didn't change anything in gui1, so it's still a blackbox wrt the patch
20101003 07:25:00< shadowmaster> *GUI1's objects
20101003 07:27:35-!- shadowm_laptop2 is now known as shadowm_laptop
20101003 07:37:22< shadowm_laptop>  /nick shadowm_tty
20101003 07:37:26< shadowm_laptop> er.
20101003 07:42:18< shadowmaster> aw.
20101003 07:42:34< shadowmaster> apparently this version of Boost.asio I have doesn't work with g++ 4.5
20101003 07:43:30< shadowmaster> or
20101003 07:43:31< shadowmaster> /usr/bin/ld: build/sconf_temp/conftest_16.o: undefined reference to symbol 'pthread_key_delete@@GLIBC_2.2.5'
20101003 07:44:33< shadowmaster> /usr/bin/ld: note: 'pthread_key_delete@@GLIBC_2.2.5' is defined in DSO /lib64/libpthread.so.0 so try adding it to the linker command line
20101003 07:44:36< shadowmaster> /lib64/libpthread.so.0: could not read symbols: Invalid operation
20101003 07:44:41< shadowmaster> how interesting.
20101003 07:46:35< Arc> there's a few c++ libraries that dont compile correctly on 4.5, i ran into this a few months ago on gentoo
20101003 07:46:58< Arc> i thought most of the problems were solved with 4.5.1 though
20101003 07:47:19< shadowmaster> I'm using 4.5.1.
20101003 07:47:44< Arc> boost doesnt compile on 4.5.1 or your app that uses boost doesnt?
20101003 07:48:18< shadowmaster> that particular Boost.asio test from the Wesnoth scons recipe doesn't work with g++ 4.5.1.
20101003 07:48:20-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 07:48:24< shadowmaster> although I guess the problem stems from Boost.asio itself
20101003 07:49:00-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 07:49:02< Arc> i dont having problems compiling boost, but i don't use it a lot since im not a c++ programmer
20101003 07:50:06< shadowmaster> I also think I just did something incredibly foolish with my system, but since it didn't break I'm glad.
20101003 07:50:06-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 07:50:11-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 07:50:21< Arc> what distro are you running?
20101003 07:50:34< shadowmaster> Debian.               
20101003 07:50:36-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 07:50:49< shadowmaster> g++ 4.5? from experimental.
20101003 07:50:56-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 07:51:07< Arc> ive found when using bleeding edge, gentoo is a lot more stable
20101003 07:51:31< shadowmaster> I'm not really in need of the bleeding edge, I was just trying this out out of curiosity ;)
20101003 07:51:56< Arc> ah well its good to try it out, imho developers should be using bleeding edge systems so they find problems early
20101003 07:52:25< shadowmaster> yeah, I wanted to check how well Wesnoth worked when compiled with 4.5
20101003 07:52:52< Arc> ill give it a shot on projects.hacdc.org
20101003 07:53:06< shadowmaster> in any case the default system compiler continues to be gcc/g++ 4.4.5. Only the libstdc++ got a forced upgrade.
20101003 07:53:16< shadowmaster> (yes, I know how to downgrade packages in case of emergency.) ;)
20101003 08:00:28< shadowmaster> silene: http://wesnoth.pastebin.com/TdGjrg9v
20101003 08:00:52< shadowmaster> not sure if there are more warnings since I wasn't paying much attention to the build.
20101003 08:00:56< shadowmaster> (g++ 4.4.5)
20101003 08:02:51< shadowmaster> I'm going to playtest LoW on trunk now. This is the first time I'll play LoW to its end since fendrin got his hands on it for 1.5.x.
20101003 08:03:23< shadowmaster> I hope to not die or get injured in the process.
20101003 08:05:46< shadowmaster> fendrin: Landar is a side 1 leader along with Kalenz in trunk, in single player mode. Is this intended?
20101003 08:06:43< silene> shadowmaster: the code is correct, the compiler seems to be confused by the fact that only part of the value is initialized (it's a 'nil' value, there is no payload)
20101003 08:07:23< shadowmaster> there's a certain GUI2 developer out there who will want to touch the code anyway to silence the warning, since he always compiles in strict mode.
20101003 08:08:18< silene> shadowmaster: this part of the code isn't supposed to be compiled in strict mode, it's c code
20101003 08:08:31< shadowmaster> oh.
20101003 08:08:45< shadowmaster> I thought the plan was to compile lua using the C++ compiler?
20101003 08:09:02< silene> shadowmaster: sure, but that doesn't change the fact that it's c code
20101003 08:09:18< silene> the build system maintainer is supposed to adjust the flags accordingly
20101003 08:11:57< shadowmaster> fendrin: also, the elvish ally in scenario 1 doesn't have the same flag style as the player. Is this intended?
20101003 08:13:17< shadowmaster> I apparently started with a bad strategy for Normal mode as well, so I'll go play HttT instead.
20101003 08:15:37-!- wesbot changed the topic of #wesnoth-dev to: 159 bugs, 298 feature requests, 14 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20101003 08:20:09-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: I've left so now you're sad.]
20101003 08:22:35-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20101003 08:22:46< mordante> servus
20101003 08:23:58< mordante> Espreon, I'm quite sure silene runs Linux
20101003 08:24:20-!- yenloc [~hhyloc@113.166.188.249] has quit [Quit: Leaving]
20101003 08:24:41< Espreon> True... Hmmm, this is odd.
20101003 08:24:56< Espreon> I wonder what version of Pango he has.
20101003 08:25:51< silene> the one shipped with the latest ubuntu, so 1.28.0 according to the package manager
20101003 08:26:47< silene> why?
20101003 08:27:26< mordante> since the tip source looks good for me with pango 1.20
20101003 08:27:45< mordante> once I update to squeeze I'll also have 1.28
20101003 08:31:09-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20101003 08:31:56< mordante> shadowm_laptop, I compile test Debian testing and there gcc 4.5 for experimental compiles Wesnoth for me
20101003 08:32:16< mordante> shadowm_laptop, at least it did a while ago, haven't updated for a few weeks
20101003 08:32:34< shadowmaster> I'm using squeeze+sid+experimental.
20101003 08:32:38< shadowmaster> but glad to kn ow I'm not the only one using compilers from exp.
20101003 08:33:22< mordante> I use squeeze and gcc form experimental (which depends on a few unstable packages)
20101003 08:33:35< mordante> from*
20101003 08:35:53-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20101003 08:37:30< mordante> shadowmaster, I haven't added strict flags to the C compiler since poolalloc caused a lot of issues and I wasn't in the mood to fix them
20101003 08:37:41< CIA-35> silene * r46886 /trunk/data/lua/location_set.lua: Added Lua library for location sets and maps.
20101003 08:37:44< CIA-35> silene * r46887 /trunk/data/lua/wml-tags.lua: Avoided costly [store_locations][or]find_in= way of adding elements to location sets.
20101003 08:37:55< shadowmaster> of course, I just didn't know that code was still compiled as C.
20101003 08:38:18< mordante> poolalloc is
20101003 08:38:58< mordante> hmm yeah the lua code is compiled with the c++ compiler
20101003 08:40:43< shadowmaster> do any of the official packagers use poolalloc anyway? why's that code still around? ISTR it caused major problems on amd64 (as in, "game unusable and randomly crashing for no reason at all")
20101003 08:41:47< mordante> no idea whether somebody uses it
20101003 08:48:28< mordante> Ivanovic, it ends with http://paste.debian.net/93002/ for me
20101003 08:48:34-!- dtiger [~dtiger@dynamic-vpdn-93-125-67-203.telecom.by] has joined #wesnoth-dev
20101003 09:06:31< CIA-35> shadowmaster * r46888 /trunk/src/terrain.hpp: Indentation one-liner (sorry, couldn't resist)
20101003 09:09:02< mordante> and thousands of compilers scream in agony ;-)
20101003 09:10:10< CIA-35> shadowmaster * r46889 /trunk/data/campaigns/Two_Brothers/scenarios/05_Epilogue.cfg: Fix bug #16811 on Two Brothers (like r46784 on Liberty)
20101003 09:10:25< shadowmaster> not those who use ccache
20101003 09:11:10< mordante> true
20101003 09:13:59-!- Blueblaze [~Blueblaze@99.148.247.46] has quit [Quit: Blueblaze]
20101003 09:18:30< shadowmaster> I'm suffering for the rest of those people anyway in order to complete a FR for eleazar
20101003 09:18:50< shadowmaster> since I'm dealing with adding actual code to terrain.hpp
20101003 09:23:43-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101003 09:45:02-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 276 seconds]
20101003 09:49:19-!- dtiger [~dtiger@dynamic-vpdn-93-125-67-203.telecom.by] has quit [Remote host closed the connection]
20101003 09:51:32-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101003 09:54:11< shadowmaster> gamemap::get_terrain_string() ?! no way
20101003 09:56:13< Espreon> Hmmmmmm?
20101003 10:02:05-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 265 seconds]
20101003 10:06:11-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20101003 10:08:36-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101003 10:28:17-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: NI]
20101003 10:30:42-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20101003 10:32:42< fendrin> hi shadowmaster
20101003 10:59:18-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20101003 10:59:58-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20101003 11:01:15-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20101003 11:01:20-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20101003 11:07:10< zookeeper> silene, actually the poor button labeling has been brought up in at least 4 different threads on the forums... no one just ever did it.
20101003 11:09:12< zookeeper> and twice someone told them it wouldn't be trivial ;)
20101003 11:10:31< silene> ...
20101003 11:11:04< silene> zookeeper: i have modified [store_reachable] to use location sets
20101003 11:11:40< zookeeper> mmh, so i see
20101003 11:19:15-!- Ivanovic [~ivanovic@dtmd-4db2d1ba.pool.mediaWays.net] has quit [Changing host]
20101003 11:19:16-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20101003 11:19:27< Ivanovic> moin
20101003 11:19:32< mordante> hi Ivanovic 
20101003 11:19:37< CIA-35> zookeeper * r46890 /trunk/data/campaigns/Two_Brothers/scenarios/03_Guarded_Castle.cfg: Fixed a broken speaker filter. Fixes bug #16810.
20101003 11:19:46< Ivanovic> mordante: perfectly normal
20101003 11:19:49< Ivanovic> those files are still created
20101003 11:20:01< Ivanovic> it is just that there will be some broken links
20101003 11:20:23< Ivanovic> and yeah, i already told the translators about that (even fixed them myself once, they rebroke with updates later on)
20101003 11:21:03< mordante> Ivanovic, ok, normally I don't expect things to work once ending with error ;-)
20101003 11:21:25< Ivanovic> what you should do is the following:
20101003 11:21:43< Ivanovic> before creating remove the files doc/manual/*html and doc/manual/*xml
20101003 11:22:08< Ivanovic> if you get some error message check if there are non empty html files for that lang
20101003 11:22:17< Ivanovic> if you get those you can clearly ignore the error
20101003 11:22:31< Ivanovic> thoug those errors are stuff in the po files
20101003 11:23:17< mordante> the rm part do you do that manually or should make do that?
20101003 11:23:45< Ivanovic> currently i have to do it manually
20101003 11:23:51< Ivanovic> but that should be done by the build system
20101003 11:24:58< mordante> ok
20101003 11:25:43< mordante> and as long as the current result is normal, I can finally start to look at make update-po4a for cmake :-)
20101003 11:26:54< Ivanovic> (the general idea is that only those files are created that are complete enough, in the current system this is decided by creating an xml file, the later steps just transform all xml files into html files, even if the files are outdated since they "exist" and if the xml files were removed, the html files for incomplete ones are not regenerated but not removed either, so before running the stuff one should remove those files)
20101003 11:27:48< mordante> ok then I'll add that to the build system
20101003 11:34:10-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20101003 11:39:55-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20101003 11:41:49< CIA-35> ilor * r46891 /branches/new_addon_server/ (9 files in 6 dirs): Merge branch 'master' into new_addon_server
20101003 11:43:16< Ivanovic> ilor: you left me some message yesterday, though i was rather busy with some guests i had
20101003 11:43:26< Ivanovic> ilor: so what is the "real" status of the new addon server?
20101003 11:43:44< Ivanovic> and have you talked to Soliton already about replacing the current server?
20101003 11:45:26< ilor> Ivanovic: it's not replace-worthy yet but at least it actually works now
20101003 11:45:59< Ivanovic> hey, this sounds like an important milestone!
20101003 11:46:00< Ivanovic> ;)
20101003 11:46:14< ilor> yep, especially since addon uploading was broken ;)
20101003 11:46:43< Ivanovic> ilor: do both ways of handling addons already work with the new server?
20101003 11:46:53< Ivanovic> that is: does addon-manager require any changes?
20101003 11:47:11< ilor> there are changes to addon-manager in the branch
20101003 11:47:28< ilor> the script generally should work though
20101003 11:47:52< Ivanovic> cool!
20101003 11:48:11< ilor> http://wiki.wesnoth.org/NewAddonServer is a status page and issue gathering place for the new addon server at the moment
20101003 11:48:43< Ivanovic> (reason for me asking this is simple since the script is what i use to remove addons and stuff like this if things are somehow borked and require some "admin work")
20101003 11:49:09< ilor> Ivanovic: some changes will be required (different auhtentication)
20101003 11:50:44< ilor> that is, if you have the script called automatically from somewhere. Adding a -L ivanovic and using a different password shouldn't be a problem
20101003 11:51:23< Ivanovic> no, i call the script by hand on my local system to access the content on the addon server
20101003 11:52:19< ilor> Soliton: around for some addon server talk?
20101003 11:52:36< Ivanovic> (that is i check the content name and email address provided to match with the one use for forum registration/messaging me and then simply call the script with the name and admin passwort to remove it)
20101003 11:53:28< ilor> Ivanovic: you'll be able to do this via the new web interface
20101003 11:53:41-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Read error: No route to host]
20101003 11:54:01< ilor> and well, umc authors should need this less since they'll be using their forum account
20101003 11:54:03< Ivanovic> aha, interesting
20101003 11:54:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20101003 11:57:37-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20101003 12:11:11-!- Bocom [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101003 12:11:16-!- Octalot [~noct@host109-157-80-24.range109-157.btcentralplus.com] has joined #wesnoth-dev
20101003 12:11:40< fendrin> wesbot: seen crab?
20101003 12:11:40< wesbot> fendrin: Sorry, I don't know of crab.
20101003 12:11:45-!- Bocom [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101003 12:11:48< fendrin> wesbot: seen crab_?
20101003 12:11:48< wesbot> fendrin: The person with the nick Crab_ last spoke 3d 14h ago.  3d 13h ago they left with the message: Quit: Leaving.
20101003 12:11:54-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20101003 12:15:31< fendrin> Ivanovic: I am going to benchmark the [tunnel] branch today. If it comes out fine, can I commit it to trunk?
20101003 12:15:57< Ivanovic> fendrin: what makes it different?
20101003 12:17:13< fendrin> Ivanovic: Well, it introduces the [tunnel] action wml tag, which let the wml coder define teleport points in a map. It can also be used to code features like [teleport] more versatile. 
20101003 12:18:21< fendrin> Ivanovic: Goes back to an old feature request from jetrel. He wanted to code forest fairies which are able to teleport in forest but only in forest.
20101003 12:19:02< Ivanovic> if it works nicely, yeah, why not commit it?
20101003 12:19:15< Ivanovic> i think it was mainly put back simply because it was done during feature freeze, right?
20101003 12:19:54< silene> Ivanovic: the thing is, i wouldn't be surprised if it made the AI 50% time slower
20101003 12:20:05< Ivanovic> woot?
20101003 12:20:14-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: need sleep]
20101003 12:20:16< Ivanovic> 50% is *extreme* especially for smaller devices
20101003 12:20:34< fendrin> Ivanovic: silene is right, it can have massive impact on ai performance. That was the reason why it was put back.
20101003 12:20:34< silene> that's why we need to benchmark it
20101003 12:20:47< Ivanovic> ah, okay
20101003 12:20:55< Ivanovic> then it makes much sense to benchmark it!
20101003 12:21:06< fendrin> Hopefully the issue can be fixed if it is present by introducing a good heuristic that can handle the teleport better.
20101003 12:29:04-!- hhyloc [~hhyloc@113.166.162.51] has joined #wesnoth-dev
20101003 12:34:58-!- GNUtoo|laptop [~gnutoo@host245-55-dynamic.180-80-r.retail.telecomitalia.it] has joined #wesnoth-dev
20101003 12:37:04-!- loonybot [~loonybot@ppp79-139-141-254.pppoe.spdop.ru] has joined #wesnoth-dev
20101003 12:37:04-!- loonybot [~loonybot@ppp79-139-141-254.pppoe.spdop.ru] has quit [Changing host]
20101003 12:37:04-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20101003 12:38:51-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20101003 12:53:11-!- yenloc [~hhyloc@113.166.162.51] has joined #wesnoth-dev
20101003 12:57:00-!- hhyloc [~hhyloc@113.166.162.51] has quit [Ping timeout: 265 seconds]
20101003 13:05:42-!- Vetinari [~lukjad@ip216-239-89-3.vif.net] has joined #wesnoth-dev
20101003 13:05:42-!- Vetinari [~lukjad@ip216-239-89-3.vif.net] has quit [Changing host]
20101003 13:05:42-!- Vetinari [~lukjad@unaffiliated/lukjad] has joined #wesnoth-dev
20101003 13:06:53< zookeeper> silene, i'm trying to apply your proximity AI patch, but all of a sudden the whole scenario has turned horribly buggy. i haven't yet found why, but i suspect some code bug since it was working fine the last time i tried...
20101003 13:07:42< silene> zookeeper: i didn't have any trouble, i just had to adjust [removeitem] into [remove_item]
20101003 13:08:22< zookeeper> i didn't do that change since 1.9.1 doesn't support that, but [removeitem] should still work
20101003 13:09:06< silene> i'm not sure it still works
20101003 13:09:39< zookeeper> territory/army labels aren't always set right, captured tile items aren't placed, turns don't always end...
20101003 13:10:02< zookeeper> ah, hmm, maybe my build is out of sync with some more recent lua changes or something.
20101003 13:11:09< zookeeper> sounds the likeliest explanation, so i guess i'll just wait for a new build and re-try then...
20101003 13:12:18< silene> zookeeper: that's strange, there hasn't been any lua changes lately; and even if there was, they shouldn't be breaking working scenarios; i've just launched a game and it's working fine for me
20101003 13:12:18-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 13:12:24-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 13:13:10-!- joo is now known as joo|sleeo
20101003 13:13:11-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 13:13:14-!- joo|sleeo is now known as joo|sleep
20101003 13:13:21-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 13:14:22-!- eleazzaar_ [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20101003 13:14:22-!- eleazzaar [~jbjerk@ppp-70-226-208-155.dsl.spfdil.ameritech.net] has quit [Read error: Connection reset by peer]
20101003 13:14:23-!- eleazzaar_ is now known as eleazzaar
20101003 13:19:35< zookeeper> silene, ok, it seems to be caused by usage of [removeitem]. _how_ that can screw up anything else than removal of items is beyond me, though.
20101003 13:20:47< silene> zookeeper: that's because it aborts the execution of the current event
20101003 13:21:00< zookeeper> ...urgh. sounds insane.
20101003 13:21:34< silene> yes, i'm not sure what was the point of this kind of backward compatibility
20101003 13:22:29< zookeeper> none? ;)
20101003 13:22:50< zookeeper> i mean, backwards compatibility is nice, i asked for it, but if it doesn't actually work then that's silly.
20101003 13:53:37< AI0867> 20101003 13:51:26 warning display: Can't verify V-Sync status.
20101003 13:53:42< AI0867> on the ogl branch
20101003 13:53:54< AI0867> results in black screen and refusal to quit
20101003 13:54:48< AI0867> alink^
20101003 13:54:57< AI0867> this is in windowed mode
20101003 13:58:01-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20101003 13:58:10< Unnheulu> SVN's bugged...
20101003 13:58:29-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Quit: Leaving]
20101003 13:59:34< Unnheulu> ie, hover the mouse over a button and it crasheds
20101003 13:59:36< Unnheulu> *crashes
20101003 14:00:11-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has joined #wesnoth-dev
20101003 14:00:11-!- Gambit [~Gambit@pa-67-234-122-243.dhcp.embarqhsd.net] has quit [Changing host]
20101003 14:00:11-!- Gambit [~Gambit@unaffiliated/gambit] has joined #wesnoth-dev
20101003 14:03:32< Ivanovic> AI0867: i always get this one
20101003 14:03:43-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20101003 14:03:51< Ivanovic> AI0867: and it works anyway (at least over here)
20101003 14:04:07< Ivanovic> what craphics hardware are you using?
20101003 14:15:37-!- wesbot changed the topic of #wesnoth-dev to: 157 bugs, 297 feature requests, 14 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20101003 14:35:11< zookeeper> mordante, isn't it entirely trivial to remove the "/transparent" profile hack from game_events.cpp?
20101003 14:35:15< AI0867> Ivanovic: i915
20101003 14:35:35< AI0867> direct rendering is functional at this time though
20101003 14:38:54-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20101003 14:47:32< zookeeper> silene, any idea why wesnoth.fire("delay", { time = 500 } ) would result in an "attempt to call field 'delay' (a nil value)" error?
20101003 14:48:25< zookeeper> maybe _that_ is due to my old build, checking...
20101003 14:49:03< zookeeper> yes, that was the cause, so nevermind.
20101003 14:55:27-!- Vetinari [~lukjad@unaffiliated/lukjad] has quit [Ping timeout: 240 seconds]
20101003 14:55:29-!- Vetinari_ [~lukjad@unaffiliated/lukjad] has joined #wesnoth-dev
20101003 15:00:56< silene> zookeeper: yes, you can just remove the [delay] from wml-tags.lua until you get a new build
20101003 15:02:16< zookeeper> yeah, that's what i did
20101003 15:07:36< zookeeper> silene, uploaded a new version of proximity with your patch. seems to work pretty well, i lost a 3p match against two AIs when playing quickly :p
20101003 15:16:57< silene> great
20101003 15:17:48< silene> i have been playing a bit more and my guess is that the ai should learn to fill the holes in order to get better
20101003 15:25:44< Ivanovic> AI0867: currently about to get the gl branch working on my laptop, too
20101003 15:31:28< silene> zookeeper: you committed the version with [remove_item]; also my latest commit should make [removeitem] backward-compatible and hence usable (pure lua change, so no need for a new build)
20101003 15:31:30< CIA-35> silene * r46892 /trunk/data/lua/location_set.lua: Added location_set:size method.
20101003 15:31:39< CIA-35> silene * r46893 /trunk/data/lua/helper.lua: Added a deprecation helper.
20101003 15:31:50< CIA-35> silene * r46894 /trunk/data/lua/wml/items.lua: Modified [removeitem] deprecation message so that it is no longer fatal.
20101003 15:33:15< Ivanovic> AI0867, alink: yeah, here with intel hardware (GM945 stuff) i get the load screen in gl branch correctly
20101003 15:33:26< Ivanovic> but then music for the main menu starts and the menu is not displayed
20101003 15:33:39< Ivanovic> (the logo from initial startup is still there)
20101003 15:33:54< Ivanovic> but when moving the mouse cursor around it does "unhide" the menu
20101003 15:34:08< Ivanovic> (using the software mouse cursor)
20101003 15:42:45< zookeeper> silene, yeah, i was digging a hole in the yard when i thought of that too...uploading a fix ;)
20101003 15:44:30< zookeeper> silene, actually, is your AI patch compatible with 1.9.1?
20101003 15:45:13< silene> zookeeper: it should be, but i can give it a try
20101003 15:45:36< zookeeper> no worry, i'm trying it out right now myself
20101003 15:47:17< zookeeper> silene, seems to work ok
20101003 15:47:28 * zookeeper goes back to digging
20101003 15:48:02< zookeeper> (i didn't bump the add-on version btw)
20101003 15:56:27< AI0867> Ivanovic: I get the load screen too, but the main menu doesn't appear
20101003 16:20:22-!- Bocom [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Read error: Connection reset by peer]
20101003 16:20:45-!- Bocom [~Bocom@c-8dcde255.013-31-6b736412.cust.bredbandsbolaget.se] has joined #wesnoth-dev
20101003 16:28:43-!- dtiger [~dtiger@dynamic-vpdn-93-125-67-43.telecom.by] has joined #wesnoth-dev
20101003 16:34:08< Ivanovic> AI0867: using mesa 7.8.2, too?
20101003 16:43:44-!- anonymissimus [~chatzilla@HSI-KBW-095-208-089-203.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20101003 16:44:34< anonymissimus> silene: just guessing; is the : lua operator s static accessor ?
20101003 16:46:00< silene> not sure what you mean, "t:f(x)" is just syntactic sugar for "t.f(t,x)"
20101003 16:46:58< silene> anonymissimus: by the way, i'm not too happy with the [condition] tag, would you mind if it was renamed to something else (especially something starting with "filter"), e.g. [filter_if]
20101003 16:47:57< anonymissimus> hm not too much; but zookeeper liked it however
20101003 16:48:19< zookeeper> [filter_condition]? :P
20101003 16:48:42< anonymissimus> I like the name too; wml beginners are confused about filters vs conditions we shouldn't mix that
20101003 16:48:53< zookeeper> [filter_if] doesn't really work, since you're not filtering for an "if"
20101003 16:49:42< zookeeper> silene, why don't you like it?
20101003 16:49:48< anonymissimus> the other idea I had in mind was to completely merge filters and conditions so that arbitrary variables are allowed in filters
20101003 16:50:20< anonymissimus> since they can be used partly for the same things; checking whether unit.id equals something etc
20101003 16:50:38< silene> zookeeper: i was suggesting [filter_if] just because its content is the same as [if], but i don't mind anyway; the reason for [filter_something] is that there is this general rule that [filter] tags are never executed; now it adds some other tags that are never executed
20101003 16:50:41< fendrin> AI0867: Hello, I fail getting crab's test script for doing the ai benchmark to run. It's the database setup that doesn't work.
20101003 16:50:47< anonymissimus> which should be done in a filter but cna now also be done in [condition]
20101003 16:52:12< anonymissimus> condition was also the name suggested on the codings page
20101003 16:53:40< anonymissimus> "static accessor": like the scope oprator :: in C++
20101003 16:54:17< zookeeper> silene, i see. well, i'm fine with [filter_condition].
20101003 16:54:39< silene> anonymissimus: no, the only thing static in lua is local variables; so ":" is closer to the virtual "." of c++
20101003 16:54:53< zookeeper> it seems a tad bit more accurate than just [condition] (but of course it's longer to type ;)
20101003 16:56:07< anonymissimus> I vote for keeping [condition] to point out the logical difference to filters
20101003 16:56:56< silene> anonymissimus: how is that different? it still filters the event
20101003 16:57:20< anonymissimus> variables aren't valid in filters
20101003 16:57:50< silene> not sure what you mean, but you can definitely write something like $x in a filter
20101003 16:58:01< anonymissimus> when browsing the wml workshop you can always see people making [if][filter] syntax
20101003 16:59:04< anonymissimus> um I mean that you can't check the values of variables being equal to something arbitrary
20101003 17:00:11< silene> i don't understand how it is an argument against the [filter_condition] tag
20101003 17:01:39< anonymissimus> well the functionality is imo very different from the other filters; it runs on variables, the other filters on units
20101003 17:02:23< anonymissimus> both things "filter" the event out in the engine but the general wml author doen't know that code
20101003 17:02:39< silene> no, the other filters run on locations and on weapons, not just on units; also it can also run on units, e.g. [have_unit]
20101003 17:03:51-!- clanehin [~quassel@cpe-174-099-073-094.nc.res.rr.com] has quit [Remote host closed the connection]
20101003 17:04:27< fendrin> Damn, the pathfind branch is faster than trunk.
20101003 17:04:34< fendrin> Something must be wrong.
20101003 17:07:09< anonymissimus> hm anyway I remember that as a wml beginner I was confused about filters vs conditions
20101003 17:07:53< anonymissimus> it is almost the same, so I wondered about merging it; that is [filter_condition] allowed inside of [filter] or such
20101003 17:09:25< anonymissimus> what is clear is that this tag adds great flexibilitym to wml at a very low cost, currently
20101003 17:10:20< anonymissimus> I rewrote some scenario, the "nesting level" is significantly reduced since you cna split up into events
20101003 17:14:11-!- yenloc [~hhyloc@113.166.162.51] has quit [Quit: Leaving]
20101003 17:21:42-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20101003 17:22:29-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20101003 17:23:14< silene> anonymissimus: i agree that it's a nice feature to have
20101003 17:25:14< zookeeper> well, [condition] is for all intents and purposes "a filter" just like [filter] is
20101003 17:25:54< zookeeper> what newbies seem to have trouble grasping is that [filter] isn't a condition
20101003 17:26:19-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20101003 17:26:19< zookeeper> (or how filtering in general works; that is, that it doesn't "pick a unit" on which following operations are applied to)
20101003 17:27:15< zookeeper> i do vaguely remember that i too was confused about it at the beginning and thought that a [filter] somehow picks a unit and that if i then do a [kill] for example, it magically gets applied to the unit which was picked by the [filter], or something like that
20101003 17:28:41-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20101003 17:39:11< anonymissimus> zookeeper: I think about it the other way: a [filter] is a condition because you could do everything that can be done with filters also with conditions about variables (I think)
20101003 17:40:24< anonymissimus> silene: however, I agree that "filters* are the only tags not being executed" is a rule worth being kept; so did we agree upon [filter_condition] ?
20101003 17:40:52< silene> yes
20101003 17:41:02< anonymissimus> (I never read about this rule in the wiki I think)
20101003 17:45:58-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20101003 17:46:29-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20101003 17:47:17-!- GNUtoo|laptop [~gnutoo@host245-55-dynamic.180-80-r.retail.telecomitalia.it] has quit [Quit: Program received signal SIGSEGV, Segmentation fault.]
20101003 17:47:26< silene> i guess i should at least write it in the description register_wml_action, since it is probably the only place where it matters
20101003 17:47:45-!- GNUtoo|laptop [~gnutoo@host245-55-dynamic.180-80-r.retail.telecomitalia.it] has joined #wesnoth-dev
20101003 17:49:22< anonymissimus> it would be helpful for wml authors in general
20101003 17:49:44< anonymissimus> on the events page there is said that filters only apply for events with units
20101003 17:50:54< anonymissimus> thats also no longer true then; condition can be in events entirely without units
20101003 17:51:37< anonymissimus> (I just see that that's another reason for why I chose condition) :P
20101003 17:53:01< silene> it was already true before, it just wasn't particularly useful (unless you wanted your event to never fire)
20101003 17:59:38-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20101003 18:00:17-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 18:01:39-!- loonybot [~loonybot@ppp79-139-141-254.pppoe.spdop.ru] has joined #wesnoth-dev
20101003 18:01:40-!- loonybot [~loonybot@ppp79-139-141-254.pppoe.spdop.ru] has quit [Changing host]
20101003 18:01:40-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20101003 18:02:29-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20101003 18:23:31< AI0867> fendrin: maybe I can give it a shot
20101003 18:24:59< AI0867> Ivanovic: 7.0.3, actually
20101003 18:25:18< Ivanovic> AI0867: ugh, okay
20101003 18:25:26< anonymissimus> silene: do you plan on improving wesnoth's SLF ? :P that library seems bit overpowered just for zookeeper's store_reachable_locations
20101003 18:26:03< AI0867> fendrin: where is the script again?
20101003 18:33:33-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20101003 18:38:45-!- fabi [~fabi@88-134-142-224-dynip.superkabel.de] has joined #wesnoth-dev
20101003 18:38:45-!- fabi [~fabi@88-134-142-224-dynip.superkabel.de] has quit [Changing host]
20101003 18:38:45-!- fabi [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20101003 18:41:22< silene> anonymissimus: not really, the library doesn't bring much more than what slf can already do; in fact the only thing it brings is the find_in= attribute of a slf, but it brings it in a way suitable for lua; so the library is primarily intended for the user (and it is documented)
20101003 18:41:45< silene> for instance, using wml variables (so that find_in= works) in a AI would be a really bad idea
20101003 18:42:38-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 264 seconds]
20101003 18:45:11-!- clanehin [~quassel@cpe-174-099-073-094.nc.res.rr.com] has joined #wesnoth-dev
20101003 19:07:41-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20101003 19:08:48< alink> AI0867: the V-Sync warning is not important. I also had it when I added it, I should disable it.
20101003 19:09:45< alink> Probably some SDL bug since they fixed things around that stuff in 1.2.14. Maybe it's still not fully fixed
20101003 19:10:51< alink> black screen is more annoying, I'll add read/write check to better isolate this
20101003 19:12:32-!- EdB [~EdB@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20101003 19:30:00< CIA-35> anonymissimus * r46895 /trunk/ (changelog data/lua/wml-tags.lua src/game_events.cpp): Renamed [event][condition] to [event][filter_condition] in order to keep the rule "The only tags inside events not being executed are the filter* ones.".
20101003 19:44:28-!- Blueblaze [~Blueblaze@adsl-99-148-247-46.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20101003 19:50:03< mordante> zookeeper, removing the hack isn't trivial, just need to find some time to make it work properly
20101003 19:53:29< Unnheulu> Svn is broken here
20101003 19:54:58< alink> Unnheulu: what do you mean? current wesboth-trunk, svn access to gna, svn on your box ?
20101003 19:55:50< Unnheulu> Wesnoth-trunk
20101003 19:56:03< Unnheulu> When I hover the mouse over a button
20101003 19:56:13< mordante> works for me Unnheulu 
20101003 19:56:33< mordante> read your message before (just finished compilation)
20101003 19:56:43< zookeeper> mordante, okay... i just don't understand how that's possible even though obviously i don't know c++.
20101003 19:56:44< Unnheulu> Ah
20101003 19:56:53< alink> Unnheulu: error message ?
20101003 19:57:20< Unnheulu> wesnoth: sdl_utils.cpp:1655: void blit_surface(const surface&, const SDL_Rect*, surface&, const SDL_Rect*): Assertion `dst_rect.x >= 0' failed
20101003 19:57:26< mordante> zookeeper, I need to properly implement the portrait tag in the unit to make it work properly
20101003 19:57:33< Unnheulu> I figured one line didn't need pastebin >.>
20101003 19:58:09< alink> Unnheulu: ah I saw this one once, we could improve this dst_rect.x restriction
20101003 19:58:22< alink> Unnheulu: you use which language in wesnoth ?
20101003 19:58:27< Unnheulu> English
20101003 19:58:43< alink> Unnheulu: screen resolution ?
20101003 19:58:43< mordante> Unnheulu, every button or some button and which resolution?
20101003 19:58:49< Unnheulu> 1920x1080
20101003 19:59:14< Unnheulu> Erm, slightly smaller than 1080
20101003 19:59:42< Unnheulu> The top three fail
20101003 19:59:52< Unnheulu> And Add-ons
20101003 20:00:48< mordante> Unnheulu, which revision and which OS do you use?
20101003 20:01:18-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20101003 20:01:19< mordante> alink, most of the time it's a bug in a higher level of the code
20101003 20:02:42< alink> yes I see now that dst_rect.x should never be negative on his setup, so improving blit_surface will just remove the assert and probably leave a misplaced tooltip
20101003 20:03:15< Unnheulu> mordante, r46895
20101003 20:04:01< alink> still, I have a patch somewhere for this restritction on blit_surface, since I also need it for ~BLIT()
20101003 20:08:29< zookeeper> mordante, if you got a minute, could you humour me and give an example of what http://wesnoth.pastebin.com/EyNBivtJ would end up regressing? i'm not going to start learning c++, but i still got interested in this particular thing ;)
20101003 20:08:59< mordante> Unnheulu, same here
20101003 20:09:21< Unnheulu> And Linux >.>
20101003 20:09:52< mordante> zookeeper, about to leave but your change will remove all transparent portraits
20101003 20:10:28< mordante> we use the same profile wml key for both the transparent and non-transparent portrait
20101003 20:10:38< zookeeper> uh, okay. so i take it that [unit_type] [portrait] image= isn't actually used, but the transparent hack always works on the [unit_type] profile= ?
20101003 20:10:38< mordante> the latter is used in the help
20101003 20:11:21< zookeeper> but ok, if there's no easy way to fix the bug of ~BLIT not working in message images then fine..
20101003 20:11:40< mordante> yeah the portrait code isn't used yet since it will fail when a unit has a custom profile
20101003 20:11:55< mordante> what's the exact bug?
20101003 20:11:55< alink> I suggest :   filename=image::locator(image).get_filename();  that will strip ~BLIT()
20101003 20:12:29< mordante> but I'm afk now
20101003 20:12:32< silene> mordante: the bug is that the last '/' is not necessarily the good one
20101003 20:12:52< silene> (and it happens whenever one uses an image path)
20101003 20:35:56-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has quit [Remote host closed the connection]
20101003 20:47:04-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 252 seconds]
20101003 20:49:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20101003 20:58:04< fabi> AI0867: utils/ai_test
20101003 20:58:12-!- fabi is now known as fendrin
20101003 21:00:51< CIA-35> alink * r46897 /branches/ogl/src/video.cpp: Also test Openg call used during scrolling.
20101003 21:01:06< CIA-35> alink * r46896 /branches/ogl/src/video.cpp: 
20101003 21:01:06< CIA-35> Do several read and write tests on front and back buffers.
20101003 21:01:06< CIA-35> Hopefully this will help to diagnose incorrect OpenGL behavior.
20101003 21:01:16< CIA-35> alink * r46898 /branches/ogl/src/video.cpp: 
20101003 21:01:16< CIA-35> Stop touching V-Sync, not even try to disable it.
20101003 21:01:16< CIA-35> because it returns a weird error (SDL bug?)
20101003 21:03:09< alink> ^people having glitches should check std output for errors. I didn't exit, so you can still run the game and check other things
20101003 21:03:46< alink> but afk now
20101003 21:04:13-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20101003 21:18:46-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20101003 21:20:18-!- phlaem [~a@212.84.227.31] has joined #wesnoth-dev
20101003 21:20:52-!- dtiger [~dtiger@dynamic-vpdn-93-125-67-43.telecom.by] has quit [Remote host closed the connection]
20101003 21:21:29-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20101003 21:25:00< Ivanovic> alink: yeah, now i get some errors
20101003 21:25:48< Ivanovic> http://pastebin.com/9vyUY2zY
20101003 21:25:53< Ivanovic> this is with gallium r600
20101003 21:28:41< CIA-35> ilor * r46899 /branches/new_addon_server/data/tools/addon_server/ (6 files in 4 dirs): NAS: clean up server code a bit, especially addon removal, partially publication, also more fixed paths removal
20101003 21:28:46< CIA-35> ilor * r46900 /branches/new_addon_server/src/addon/manager.cpp: fix wrong order of arguments in call to install_addon
20101003 21:28:52< CIA-35> ilor * r46901 /branches/new_addon_server/data/tools/wesnoth_addon_manager: very crude 500 error reporting in script publish
20101003 21:30:19< Ivanovic> alink: but nothing is shown for the intel hardware
20101003 21:31:12< Ivanovic> alink: pasted this stuff for the radeon devs to have a look at (that is the log output as well as your test code since that appears to be as basic a test as it can get)
20101003 21:31:39< silene> Ivanovic: you can tell them that it works fine with fglrx
20101003 21:31:46< silene> (if you haven't already done so)
20101003 21:31:52< Ivanovic> silene: really, does it?
20101003 21:32:02< Ivanovic> silene: when i used fglrx i had render artifacts ingame!
20101003 21:32:09< Ivanovic> kind of "image ghosting"
20101003 21:32:21< silene> Ivanovic: yes, there are some artifacts, but nothing major
20101003 21:32:26< Ivanovic> for me r600 classic appears to "always" work
20101003 21:32:45< Ivanovic> r600g had some issues before in the load screen area and is no, using 2.6.36-rc having serious issues
20101003 21:32:57< Ivanovic> s/no,/now,
20101003 21:33:29< Ivanovic> mesa software (LIBGL_ALWAYS_INDIRECT=1) appears to be working nicely, too
20101003 21:33:42-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20101003 21:33:53< Ivanovic> though it does not on my laptop with intel hardware, there i always get nothing displayed correctly after the initial load screen
20101003 21:34:06< silene> i wouldn't be surprised if most of the glitches i'm experiencing are due to reading back the alpha channel of the backbuffer
20101003 21:38:56< Ivanovic> yeah, could be
20101003 21:42:15-!- Mica__ [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101003 21:42:28-!- Unnheulu [~ieuan@cpc1-pnth2-2-0-cust788.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20101003 21:43:27-!- Mica__ is now known as Mica
20101003 21:43:39< CIA-35> ivanovic * r46902 /trunk/po/ (wesnoth/vi.po wesnoth-sotbe/vi.po wesnoth-units/vi.po): updated Vietnamese translation
20101003 21:43:44< CIA-35> ivanovic * r46903 /branches/1.8/ (5 files in 4 dirs): updated Vietnamese translation
20101003 21:47:55-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20101003 21:53:02< CIA-35> ivanovic * r46904 /branches/1.8/ (24 files in 2 dirs): 
20101003 21:53:02< CIA-35> fix Indonesian and Chinese (Traditional) manuals so that they can create correct html files
20101003 21:53:02< CIA-35> regenerated doc files
20101003 21:57:48< CIA-35> ivanovic * r46905 /trunk/po/wesnoth-manual/ (id.po zh_TW.po): fixed Indonesian and Chinese (Traditional) manuals so that they can create correct html files
20101003 21:58:09< Ivanovic> mordante: the files in trunk should now create correct stuff
20101003 21:58:26< Ivanovic> running a pot-update (and make update-po4a) in trunk now so that you get "up to date" files
20101003 21:59:34-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has joined #wesnoth-dev
20101003 22:02:46< mordante> silene, not entirely sure what you mean
20101003 22:04:00< Ivanovic> okay, this pot-update appears to be lovely...
20101003 22:05:08< CIA-35> ivanovic * r46906 /trunk/ (1083 files in 25 dirs): 
20101003 22:05:08< CIA-35> pot-update
20101003 22:05:08< CIA-35> regenerated doc files
20101003 22:06:44< CIA-35> ivanovic * r46907 /trunk/data/core/units/drakes/Flameheart.cfg: fix a forgotten apostroph conversion...
20101003 22:07:28< silene> mordante: imagine that the image is "a.png~X(/)", it gets translated into "a.png~X(/transparent/)"; then the engine dies because X doesn't expect that
20101003 22:09:02< CIA-35> ivanovic * r46908 /trunk/po/wesnoth-units/ (52 files): apply pofix.py to the -units textdomain
20101003 22:10:19< mordante> silene, ok, at least that doesn't seem to be used yet, but I still want to rewrite that code
20101003 22:10:53< silene> mordante: what do you mean that doesn't seem used to? zookeeper has been complaining about it for months; if it was working, it would be using it
20101003 22:15:03< mordante> he mentioned some problem a few weeks ago which had nothing the do with the / but with the adding of /transparent
20101003 22:16:35< silene> mordante: you are missing point, they are the same problem; the issue is that transparent is added at the wrong place; if the code was producing "transparent/a.png~X(/)", everything would be fine
20101003 22:17:19< mordante> is that^ true zookeeper ?
20101003 22:18:57< CIA-35> ilor * r46909 /branches/new_addon_server/data/tools/addon_server/addons/views.py: NAS: fix issue with invalid directory addons were put in, handle being unable to delete the old file
20101003 22:25:15< mordante> Ivanovic, indeed automake can now run update-po4a without errors :-)
20101003 22:25:26< mordante> so good base to look at the cmake version
20101003 22:25:59< SpoOkyMagician> just got an error in a replay. (gathering some info first though)
20101003 22:26:59< SpoOkyMagician> ok so stderr.txt was no help...
20101003 22:30:31< SpoOkyMagician> the error: (http://img842.imageshack.us/img842/779/replayerror.png)
20101003 22:30:59< SpoOkyMagician> wesnoth 1.8.5
20101003 22:34:09-!- oye [~oye@90.169.172.234] has quit [Changing host]
20101003 22:34:09-!- oye [~oye@unaffiliated/oye] has joined #wesnoth-dev
20101003 22:35:54-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20101003 22:37:20< SpoOkyMagician> it seems to happen when i "stop" the replay (in a complete play though of the replay)
20101003 22:37:58< SpoOkyMagician> (ill try other replays to be sure)
20101003 22:39:42< mordante> I'm off night
20101003 22:40:00-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20101003 22:40:46< SpoOkyMagician> well, that one crashed wesnoth... (no error)
20101003 22:41:30 * SpoOkyMagician shrugs
20101003 22:42:00< SpoOkyMagician> i'll file a bug report I suppose.
20101003 22:45:55-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 240 seconds]
20101003 22:46:33-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20101003 22:51:10< Ivanovic> alink: just got a reply from glisse (a mesa dev)
20101003 22:51:20< SpoOkyMagician> https://gna.org/bugs/index.php?16815
20101003 22:51:31< SpoOkyMagician> I hope this helps
20101003 22:51:41< SpoOkyMagician> (and is enough information)
20101003 22:58:14< Ivanovic> alink: can you make the tests you do into a seperate tiny app?
20101003 22:58:23< Ivanovic> alink: airlied in #radeon just asked for this
20101003 22:59:32-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has quit [Quit: Be back later. :)]
20101003 23:00:11< Ivanovic> alink: the relevant parts from the chat in #radeon : http://pastebin.com/Jdm9AZrq
20101003 23:01:55< alink> back
20101003 23:02:05< alink> Ivanovic: ok reading
20101003 23:02:16< Ivanovic> alink: you might want to join #radeon, too
20101003 23:02:36< Ivanovic> (you know, those radeon mesa devs might have some good ideas about how things are supposed to work and how not)
20101003 23:02:48-!- Gambit [~Gambit@unaffiliated/gambit] has quit [Changing host]
20101003 23:02:48-!- Gambit [~Gambit@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20101003 23:04:28< alink> yeah, I guessed that my front buffer trick used to emulate SDL was non-standard, but your logs indicates that back buffer doesn't work too
20101003 23:05:12< Ivanovic> alink: i'd say that talking to the mesa devs could be a good idea to get to know what is a miractle if it actually works
20101003 23:05:12< alink> btw, all test fails :-( I could have just write the first one 
20101003 23:05:14< Ivanovic> ;)
20101003 23:06:09< alink> ok I will write a short standalone and talk to them
20101003 23:06:31< alink> Ivanovic: I might need your help again for testing it, since all works fine here
20101003 23:06:42< Ivanovic> ain't no problem
20101003 23:08:06< alink> and if this driver thing cost too much work, I may just ignore it and hurry to rewrite all code for normal double-buffering
20101003 23:08:21< Ivanovic> :)
20101003 23:08:51< SpoOkyMagician> (also found another error in similar testing. (updated bug post)
20101003 23:09:51< alink> that SDL emulation trick is useful for the transition, but it's temporary(even if possibly for a long time) so no need to polish it to much
20101003 23:10:22-!- shadowm_laptop is now known as shadowm_nolights
20101003 23:11:12-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20101003 23:16:30-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20101003 23:17:51< CIA-35> alink * r46910 /branches/ogl/src/sdl_utils.cpp: Quick blind shot attempting to fix "ghosting" with some driver
20101003 23:20:02-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20101003 23:34:16-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 252 seconds]
20101003 23:35:42< zookeeper> mordante, yes, that's true as far as i can tell
20101003 23:37:05-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: night all]
20101003 23:38:12< alink> mordante: yes, indeed, even if relatively new since it's specific to ~BLIT() because it's the first one having url as parameter
20101003 23:38:27< alink> like:   a.png~BLIT(dir/b.png)
20101003 23:39:00< alink> just need to be careful when handling image url having ~IPF
20101003 23:39:40< alink> but was already the case in the past with other stuff into the parenthesis like comma
20101003 23:39:41< Ivanovic> okay, r46910 makes no difference on my intel system
20101003 23:39:53< Ivanovic> (and yeah, there are no error messages there either)
20101003 23:40:18< alink> yeah, as I said just blind shot
20101003 23:41:14< alink> in theory, disabling blending is faster too, but I don't think that hardware has any real cost for that these days
20101003 23:41:41< alink> Ivanovic: they might be right with the glFlush thing, I will add it to the test 
20101003 23:42:27< Ivanovic> alink: but as you heard, if flushing really fixes it, it is also a driver issue
20101003 23:44:17-!- shadowm_nolights is now known as shadowm_laptop
20101003 23:52:37< CIA-35> alink * r46911 /branches/ogl/src/video.cpp: When failed, retry framebuffer tests using glFlush
20101003 23:54:17< alink> done, I think I see another possible implementation for the SDL emulator. But I am really tired about this. I go back to the real stuff: textures
20101003 23:56:23< Ivanovic> http://pastebin.com/Z1avF3T4
20101003 23:57:32< alink> Almost a goot thing, all these glFlush would kill the perf of this already slow hack
20101003 23:57:44< Ivanovic> :)
20101003 23:57:45< alink> s/goot/good
20101003 23:58:52-!- Mica [~Mica@d27-96-76-1.nap.wideopenwest.com] has joined #wesnoth-dev
20101003 23:59:04< Ivanovic> and yeah, better concentrate on the "real" stuff anyway
--- Log closed Mon Oct 04 00:00:31 2010