--- Log opened Sun Jul 04 00:00:44 2010
20100704 00:07:46-!- timotei [~Timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100704 00:28:31-!- Blarumyrran [~Blarumyrr@unaffiliated/blarumyrran] has quit [Read error: Connection reset by peer]
20100704 00:42:22-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100704 00:50:49-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100704 00:58:08-!- ElCaballero [~saul@201.243.19.27] has quit [Ping timeout: 260 seconds]
20100704 01:04:12< Elvish_Pillager> In animations, can you use [if] [else] for anything other than the hits= attribute?
20100704 01:07:36-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Quit: Got to go, I'm in a hurry so bye.]
20100704 01:15:51-!- ElCaballero [~saul@190.202.248.23] has joined #wesnoth-dev
20100704 01:18:53-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100704 01:22:57-!- ElCaballero [~saul@190.202.248.23] has quit [Ping timeout: 240 seconds]
20100704 01:23:46-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100704 01:24:39-!- DesertPanther_ [~Khalid@41.131.120.107] has quit [Quit: Leaving]
20100704 01:34:33-!- billynux [ba6c9402@wesnoth/developer/billynux] has joined #wesnoth-dev
20100704 01:37:38< billynux> People, I'm getting "Window not defined" in settings.cpp:283 from the latest revision :(
20100704 01:37:49< billynux> Did the last commit break something?
20100704 01:42:01< Espreon> Who knows?
20100704 01:44:12< billynux> ...
20100704 01:44:36-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100704 01:44:46-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Client Quit]
20100704 01:48:08-!- ElCaballero [~saul@201.243.19.27] has joined #wesnoth-dev
20100704 02:21:45-!- billynux [ba6c9402@wesnoth/developer/billynux] has quit [Quit: Page closed]
20100704 02:48:09-!- ilor [~user@wesnoth/developer/ilor] has quit [Ping timeout: 260 seconds]
20100704 02:50:58-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100704 02:51:43-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100704 03:22:47< Aethaeryn> wesbot: seen crab_
20100704 03:22:47< wesbot> Aethaeryn: The person with the nick Crab_ last spoke 1d 1h ago.  1d 1h ago they left with the message: Quit: Leaving.
20100704 03:22:51< Aethaeryn> :-(
20100704 03:34:43-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100704 03:43:49-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 264 seconds]
20100704 03:45:29-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100704 03:51:43-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: rebooting]
20100704 03:56:26-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 04:02:17-!- shadowm_tty [ignacio@wesnoth/developer/shadowmaster] has quit []
20100704 04:02:23-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 04:09:34-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: hibernating]
20100704 04:12:20-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 04:30:16-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Quit: I hope y'all have a nice day ;)]
20100704 04:38:53-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: Lost terminal]
20100704 04:39:14-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 04:45:14-!- hhyloc [~hhyloc@113.166.170.167] has joined #wesnoth-dev
20100704 04:48:11-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: QED]
20100704 05:03:05-!- Huynh [~hhyloc@113.166.170.167] has joined #wesnoth-dev
20100704 05:04:37-!- elvish_sovereign [~Zoltan@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100704 05:06:32-!- hhyloc [~hhyloc@113.166.170.167] has quit [Ping timeout: 248 seconds]
20100704 05:23:08-!- un214 [~quassel@adsl-75-45-4-19.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100704 05:30:36-!- Gambit [~quassel@67.234.73.7] has quit [Quit: Aint the fire inside? Let's all go stand around it.]
20100704 05:32:29-!- un214 [~quassel@adsl-75-45-4-19.dsl.scrm01.sbcglobal.net] has quit [Remote host closed the connection]
20100704 05:48:50-!- Truongan [~pntruonga@115.75.70.58] has joined #wesnoth-dev
20100704 05:49:00-!- Truongan [~pntruonga@115.75.70.58] has left #wesnoth-dev []
20100704 05:50:40-!- elvish_sovereign [~Zoltan@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Quit: elvish_sovereign]
20100704 05:58:52-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100704 05:59:04< mordante> servus
20100704 05:59:12< mordante> gabba, around?
20100704 05:59:54< gabba> yup mordante, hi
20100704 06:01:24< shadowmaster> mordante: no, I wasn't looking for you. I just wanted to know for how long you'd been away since I was missing you somewhat ;)
20100704 06:01:43< mordante> regarding r43834 I'm not so happy with it it's ugly I rather remove that warning as default or when somebody sets NDEBUG s/he should manually remove the warning
20100704 06:02:09< mordante> shadowmaster, ah ok well was rather busy and irc is somewhat distracting at times
20100704 06:02:53< shadowmaster> yeah, I know. Good that irssi won't send me visual notifications anyway and I can just minimize the terminal emulator and forget about IRC :P
20100704 06:03:08< gabba> mordante: I agree it's ugly, but I propose a third solution: avoid using asserts like that
20100704 06:03:12< shadowmaster> (or disconnect the SSH session)
20100704 06:03:46< gabba> mordante: asserts were always meant as a temporary solution, and people tend to forget that they disappear in release builds
20100704 06:04:10< mordante> shadowmaster, yes but I still look at times
20100704 06:04:31< mordante> gabba, we don't disable the asserts in release builds
20100704 06:04:57< mordante> and I like to use them a lot since they help tracking invalid states
20100704 06:05:01< gabba> mordante: ok, didn't know there was a compiler setting for that -- good thing
20100704 06:05:37< gabba> but disabling the warning manually every time one wants to make a release build is very unwieldly, especially for new devs
20100704 06:06:34< mordante> what compiler/build system do you use?
20100704 06:06:39< gabba> cmake
20100704 06:07:51< mordante> ok, but then we should fix the default cmake settings
20100704 06:08:25 * shadowmaster notes that asserts in release builds are occasionally good for discovering regressions.
20100704 06:08:52< gabba> mordante: really since the problem only comes up in a dozen places in the code it's easier to write the code a bit differently than change all the build systems, I think
20100704 06:08:58< gabba> but it's not my call
20100704 06:09:11 * shadowmaster . O o (also for detecting pointers to NULL objects before they propagate around and become harder to track)
20100704 06:09:31< mordante> cmake is the only system that by default adds the NDEBUG define
20100704 06:10:06< mordante> and I don't want to have these ugly things in the code (besides I will add new ones (not on purpose but it will happen))
20100704 06:10:19< gabba> mordante: I see
20100704 06:12:40< mordante> the other problem I see ones a variable gets used the comment remains and is not longer valid leading to new confusion
20100704 06:13:54< mordante> I think we can do some places better eg instead of const twindow* = get_window(); assert(window); use assert(get_window());
20100704 06:14:33< mordante> and the throw at the end of the widget/helper.cpp is a good thing
20100704 06:14:40< gabba> yes, I did a change of that type at at least one place
20100704 06:15:21< gabba> The throw is probably not well done :P, but I hesitated to add an #include <exception> 
20100704 06:16:04< gabba> no idea what prints in the console when you throw a char array
20100704 06:17:08< gabba> mordante: so NDEBUG only controls assert, it has nothing to do with the inclusion of debugging symbols?
20100704 06:17:09< mordante> #include <exception> int main() { throw "What happens?; } ;-)
20100704 06:17:30< gabba> mordante: true, shame on my laziness
20100704 06:18:31< mordante> yes NDEBUG only changes what assert does (IIRC nothing else in the library does anything with it)
20100704 06:19:53< mordante> I'll try to fix cmake next week so it adds the -Wno-unused-parameters flag in release mode
20100704 06:21:22< mordante> after that I like to review and partly revert the patch
20100704 06:23:23< gabba> mordante: ok, no objections. And at that time I can do the revert if you want, after we discuss what to do with the various cases.
20100704 06:24:03< mordante> personally I like to do the following:
20100704 06:24:08< mordante> keep the throw
20100704 06:24:13< gabba> Yeah, so throwing a char array prints "terminate called after throwing an instance of 'char const*'", it's utterly useless :P
20100704 06:24:40< gabba> we'd need a standard exception
20100704 06:25:13< mordante> there might be one, not sure or it's at my work where I use that
20100704 06:25:58< mordante> I like to change the functions calls that result in an unused variable to be put directly in the assert
20100704 06:26:35< mordante> and revert the rest, or alternatively wrap them in a marco that only does something if NDEBUG is defined
20100704 06:26:57< mordante> but that might result in broken builds once the variable is removed so that's not really my preference
20100704 06:31:05< gabba> #include <stdexcept> int main() { throw std::runtime_error("ouch"); }
20100704 06:31:17< gabba> ^this works and prints the following two lines:
20100704 06:31:18< gabba> terminate called after throwing an instance of 'std::runtime_error'
20100704 06:31:18< gabba>   what():  ouch
20100704 06:31:53< gabba> I'm surprised that it doesn't print the line number like asserts, I'm too used to Java exceptions I guess
20100704 06:34:49< mordante> no only asserts do that by default, but you can add __LINE__ to do so 
20100704 06:35:40< mordante> and it might not hurt to add the file and line info in that error
20100704 06:44:22-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100704 06:45:17< gabba> remotely related question: how comes in wesnoth code the + operator concatenates strings? I don't think I ever saw working in C++. Did we redefine the + operator for std::basic_string somewhere?
20100704 06:47:05< mordante> no it's standard http://www.cppreference.com/wiki/string/string_operators
20100704 06:47:56< mordante> what won't work in c++ is "foo" + "bar" maybe you mixed that up
20100704 06:48:24< mordante> but just for fun "foo" "bar" does work ;-)
20100704 06:49:54< shadowmaster> std:::string("foo") +  std::string("bar") should also work
20100704 06:50:13< CIA-87> mordante * r43868 /trunk/src/ana/api/timers.hpp: Add a missing include.
20100704 06:50:17< CIA-87> mordante * r43869 /trunk/src/hash.cpp: Converted the old style casts to c++ style casts.
20100704 06:50:20< shadowmaster> i.e. it'll work as long as one of the operators is unmistakeably a basic_string
20100704 06:50:22< CIA-87> mordante * r43870 /trunk/src/md5.cpp: Converted the old style casts to c++ style casts.
20100704 06:50:24< CIA-87> mordante * r43871 /trunk/src/md5.hpp: Fix invalid assumptions about integer sizes.
20100704 06:50:27< shadowmaster> ...
20100704 06:50:51< mordante> std::string("foo") + "bar" as well
20100704 06:50:59< shadowmaster> yep
20100704 06:51:10< gabba> hmm, can't find the place but I think I also saw std::string + int concatenation working, which is maybe what surprised me
20100704 06:51:24< mordante> missed your messages between my commits ;-)
20100704 06:51:43< shadowmaster> mordante: yeah, I expected that; hence the "..."
20100704 06:52:36< mordante> ah ok I already wondered what the ... was for ;-)
20100704 06:52:41< gabba> well thanks it's a good reminder of how strings work in c++, I tend to mix up some details when switching languages
20100704 06:54:15< mordante> I know that feeling ;-)
20100704 06:54:45< mordante> I stopped counting the number of times I typed bool in C
20100704 06:55:22< shadowmaster> some C standard includes a bool type generally implemented with ints, IIRC
20100704 06:55:43< shadowmaster> maybe C99
20100704 06:55:47< mordante> I need to code in C89 (embedded compiler)
20100704 06:56:36< shadowmaster> ah :(
20100704 06:57:43-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: switching kernels]
20100704 06:59:14-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100704 07:00:05-!- ancestral [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has joined #wesnoth-dev
20100704 07:04:37-!- Huynh [~hhyloc@113.166.170.167] has quit [Quit: Quit]
20100704 07:05:13-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 07:06:29-!- ancestral [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has quit [Ping timeout: 245 seconds]
20100704 07:09:15-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20100704 07:11:20-!- ancestral [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has joined #wesnoth-dev
20100704 07:12:44-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 07:13:48-!- ancestral_ [~ancestral@166.137.137.66] has joined #wesnoth-dev
20100704 07:15:56-!- ancestral_ [~ancestral@166.137.137.66] has quit [Remote host closed the connection]
20100704 07:25:39-!- ancestral [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has quit [Ping timeout: 245 seconds]
20100704 07:28:33-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has joined #wesnoth-dev
20100704 07:29:18-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100704 07:32:03-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has quit [Changing host]
20100704 07:32:03-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100704 07:42:37-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20100704 08:04:06-!- ElCaballero [~saul@201.243.19.27] has quit [Quit: Leaving.]
20100704 08:31:34-!- ancestral_ [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has joined #wesnoth-dev
20100704 08:37:11-!- ancestral_ [~ancestral@c-98-213-154-3.hsd1.il.comcast.net] has quit [Ping timeout: 265 seconds]
20100704 09:11:25< Ivanovic> moin
20100704 09:11:37< mordante> hi Ivanovic 
20100704 09:13:27-!- Appleman1234 [~Appleman1@CPE-60-226-178-176.qld.bigpond.net.au] has joined #wesnoth-dev
20100704 09:15:30< CIA-87> ivanovic * r43872 /trunk/po/wesnoth-l/vi.po: updated Vietnamese translation
20100704 09:15:33< CIA-87> ivanovic * r43873 /branches/1.8/po/wesnoth-l/vi.po: updated Vietnamese translation
20100704 09:21:47-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100704 09:35:15-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100704 09:37:35< Crab_> Aethaeryn: hello. you've looked for me ?
20100704 09:41:41< Aethaeryn> yes
20100704 09:45:32< Aethaeryn> A while back you recommended some C++ books?
20100704 09:47:53 * Crab_ is looking in the logs...
20100704 09:48:13< Crab_> well, you can buy/get something like http://www.stroustrup.com/Programming/      also, see http://www.codeguru.com/cpp/tic/      also, see http://www.cplusplus.com/doc/tutorial/
20100704 09:49:21< mordante> Crab_, did you read "Programming -- Principles and Practice Using C++" ?
20100704 09:50:06< Crab_> afair, yes
20100704 09:51:50< mordante> ok I'm curious how good it is (if somebody asks me what a good book is), but I don't want to buy it
20100704 09:54:58< Crab_> well, it's not as good as some python books I've recently read, but it's definitely readable )
20100704 09:56:27< mordante> thanks
20100704 09:58:32< CIA-87> gabba * r43874 /trunk/src/whiteboard/ (manager.cpp manager.hpp): Whiteboard: made scoped_real_unit_map more useful by making it remember the real/planned unit map status when it's created. Also some extra checks in planned_unit_map.
20100704 09:58:40< CIA-87> gabba * r43875 /trunk/src/ (3 files in 2 dirs): Whiteboard: Limit move creation arrows to maximum move of the unit. Doesn't yet take into account previously defined moves for this same unit.
20100704 09:58:46< CIA-87> gabba * r43876 /trunk/src/whiteboard/side_actions.cpp: Whiteboard: Ensure action queue is never modified while the planned unit map is active.
20100704 09:58:55< CIA-87> gabba * r43877 /trunk/src/ (12 files in 2 dirs): Whiteboard: Modify the system for applying temp modifiers on the unit map, to prepare for movement modifications.
20100704 09:59:04< CIA-87> gabba * r43878 /trunk/src/whiteboard/mapbuilder_visitor.cpp: Whiteboard: ensure an action is valid before using it to build the temp unit map.
20100704 09:59:12< CIA-87> gabba * r43879 /trunk/src/whiteboard/ (move.cpp move.hpp validate_visitor.cpp): Whiteboard: update movement points for units when building the future unit map. Doesn't take into account some stuff such as village capturing yet.
20100704 10:04:25< Crab_> mordante: looked at your change to aspect_attacks.cpp - all ok, in fact, even more code is not used and can be removed from there.
20100704 10:06:36< mordante> Crab_, ok good
20100704 10:06:49< mordante> I assume you'll axe that code
20100704 10:06:56< Crab_> yes
20100704 10:07:08-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: ZzzZzZZzzzZzzZzzzZzzzz]
20100704 10:07:44< mordante> ok
20100704 10:07:58< Crab_> you used a static analysis tool to find the code which can be removed ?
20100704 10:08:23< Aethaeryn> Crab_: okay, ty
20100704 10:09:42< CIA-87> crab * r43880 /trunk/src/ai/testing/aspect_attacks.cpp: remove unnecessary code
20100704 10:09:48< mordante> Crab_, yes ccpcheck, which I run on a regular basis
20100704 10:10:08< Crab_> it's strange that that tool hasn't found ^
20100704 10:11:27< Crab_> strictly speaking, the code you removed has changed the best_results, so, if the tool hasn't considered 'best_results' to be unnecessary, then it was wrong to suggest the removal of the changes you've done ;)
20100704 10:12:50< mordante> wouldn't be the first bug in it ;-)
20100704 10:13:00-!- ilor [~user@wesnoth/developer/ilor] has joined #wesnoth-dev
20100704 10:13:05< mordante> and it often finds more things after a rerun
20100704 10:13:45< mordante> too bad the program got much slower recently :-( so can't run it as often as before
20100704 10:14:38< Crab_> is it because of the program? or because of wesnoth ?
20100704 10:16:15< mordante> the program after some git update it just took much longer
20100704 10:16:38< mordante> of course we also add code, but I don't think we added that much
20100704 10:16:50< mordante> but I didn't look closer into it
20100704 10:32:55-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100704 10:45:31-!- boucman [c1f81d26@wesnoth/developer/boucman] has joined #wesnoth-dev
20100704 11:23:13-!- boucman [c1f81d26@wesnoth/developer/boucman] has quit [Quit: Page closed]
20100704 11:24:45< CIA-87> gabba * r43881 /trunk/src/whiteboard/ (manager.cpp manager.hpp move.cpp): Whiteboard: added custom deleter to automatize removing fake units from display on deletion.
20100704 11:49:02-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100704 11:49:27-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100704 11:49:57-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100704 11:52:41< Ivanovic> zookeeper: what is the current status of your backporting of the balancing changes?
20100704 11:53:56-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100704 12:07:16< Upthorn> hrm, all of a sudden md5.cpp is spewing out a bunch of compiler errors at me.
20100704 12:07:32< Crab_> paste them somewhere 
20100704 12:07:45-!- timotei [~Timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100704 12:08:09< Upthorn> http://wesnoth.pastebin.com/JfK2yuyx
20100704 12:08:55< Upthorn> err, http://wesnoth.pastebin.com/wUH8KPSM without erroneous syntax highlighting
20100704 12:09:58< Crab_> can you try to diff your project file with the 'last known to work' ?
20100704 12:11:04< Crab_> wesbot: log 43871
20100704 12:11:06< wesbot> mordante * r43871 : Fix invalid assumptions about integer sizes.
20100704 12:11:06< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=43871
20100704 12:11:13< Crab_> Upthorn: see ^
20100704 12:13:22-!- hhyloc [~hhyloc@113.166.148.160] has joined #wesnoth-dev
20100704 12:13:42< Crab_> Upthorn: I think that the solution is to use uint4, uint2, uint1 in all the signatures, both in .hpp and .cpp file
20100704 12:14:24< Crab_> mordante: around ?
20100704 12:15:42< Crab_> Upthorn: looks like the .cpp file is not updated properly after header changes in r43871 , so it's kind of incompatible with it
20100704 12:16:18< Crab_> Upthorn: so you can either fix the .cpp, or restore the old .hpp for you, until someone else sorts it out.
20100704 12:27:33-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100704 12:28:28< zookeeper> the current status is that i'll get to it right away
20100704 12:32:07< zookeeper> Ivanovic, ^ (funny, i was pretty sure i typed iva<tab> there...)
20100704 12:32:28< Ivanovic> :)
20100704 12:38:02-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-76.telecom.by] has joined #wesnoth-dev
20100704 12:38:28< CIA-87> zookeeper * r43882 /trunk/data/campaigns/Liberty/scenarios/08_Glory.cfg: Fixed a dialogue glitch if Dommel is killed by the orcs.
20100704 12:39:19-!- Huynh [~hhyloc@113.166.148.160] has joined #wesnoth-dev
20100704 12:39:56-!- Huynh [~hhyloc@113.166.148.160] has quit [Client Quit]
20100704 12:41:34< CIA-87> zookeeper * r43883 /trunk/data/campaigns/Son_Of_The_Black_Eye/scenarios/06_Black_Flag.cfg: Upped the turn limit from 30/25/20 to 30/28/24.
20100704 12:42:44-!- hhyloc [~hhyloc@113.166.148.160] has quit [Ping timeout: 240 seconds]
20100704 12:43:29< Upthorn> hrm
20100704 12:43:40< Upthorn> I am now getting a linker error about libboost_date_time-vc90-mt-1_42.lib
20100704 12:44:05-!- Gambit [~quassel@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100704 12:44:14< Crab_> the same which timotei has got, isn't it ?
20100704 12:44:47< Upthorn> yes, except that I never got this error before
20100704 12:45:24< Crab_> can you paste the errors ?
20100704 12:46:00< Upthorn> 1>LINK : fatal error LNK1104: cannot open file 'libboost_date_time-vc90-mt-1_42.lib'
20100704 12:47:05< Upthorn> downloading from sourceforge
20100704 12:47:43< CIA-87> zookeeper * r43884 /trunk/data/campaigns/The_Rise_Of_Wesnoth/scenarios/03_A_Harrowing_Escape.cfg: Fixed a broken speaker filter.
20100704 12:48:09-!- timotei21 [~Timotei@193.34.191.4] has joined #wesnoth-dev
20100704 12:48:13< Upthorn> seems to have done the trick.
20100704 12:48:44< Crab_> it's still interesting 'why?'  - i.e. , what part of wesnoth requires that nowadays..
20100704 12:48:56< timotei21> hey
20100704 12:49:00< Upthorn> I guess that timotei added something
20100704 12:49:05< timotei21> Upthorn: I can give you the lib :D
20100704 12:49:12< Upthorn> sourceforge gave me the lib
20100704 12:49:23< Upthorn> but thanks for the offer
20100704 12:49:24< timotei21> that's why I told Crab_ he can update the .zip file
20100704 12:49:27< timotei21> oh ok
20100704 12:49:45< timotei21> Upthorn: btw, didn't you had also some issues on older glibs?
20100704 12:49:46< Crab_> yes, I will update it. I think that it's the asio code...
20100704 12:49:48< Upthorn> but yes, it would probably be good to update the externals.zip, though
20100704 12:50:08< timotei21> Crab_: I needed to update some other libraries in the "zip" file to compile aswell
20100704 12:50:16< Crab_> which ones ?
20100704 12:50:20< timotei21> one second
20100704 12:50:24< Upthorn> timotei: either I am not sure what you mean or you have got me confused with someone else
20100704 12:50:33-!- timotei [~Timotei@wesnoth/developer/timotei] has quit [Ping timeout: 260 seconds]
20100704 12:51:05< timotei21> there were missing some ... header files: from glib
20100704 12:51:18< Upthorn> I don't remember having that problem
20100704 12:51:24< timotei21> I've downloaded the verstion: 2.22.4 and it worked
20100704 12:51:33< timotei21> there was "gcache" or something like that missing
20100704 12:51:35-!- timotei21 is now known as timotei
20100704 12:51:40< Upthorn> but there was a second zip that somebody here gave me
20100704 12:51:43-!- timotei [~Timotei@193.34.191.4] has quit [Changing host]
20100704 12:51:43-!- timotei [~Timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100704 12:51:44< Crab_> ok, thanks
20100704 12:51:48< Upthorn> so it's possible that it compensated
20100704 12:51:56< timotei> Upthorn: I think so
20100704 12:52:16< timotei> Crab_: do you want  me to send you that .lib for date-time
20100704 12:52:17< timotei> ?
20100704 12:52:34< Crab_> no, thanks
20100704 12:52:36< timotei> ok
20100704 12:53:12< Upthorn> http://sourceforge.net/projects/boost/files/ has the libs available, already built, anyway
20100704 12:53:39< Upthorn> you just have to clikc on boost-binaries and find the one you want in there
20100704 12:54:04< timotei> wow, didn't know they have already them built
20100704 13:02:42< CIA-87> zookeeper * r43885 /trunk/data/campaigns/The_Rise_Of_Wesnoth/scenarios/ (9 files): Used more varied portraits for the orcish leaders in TRoW.
20100704 13:18:36< CIA-87> zookeeper * r43886 /trunk/data/campaigns/Descent_Into_Darkness/ (maps/10_Alone_at_Last.map scenarios/10_Alone_at_Last.cfg): 
20100704 13:18:36< CIA-87> Made 'Alone at Last' somewhat easier by making the paladin side appear on turn
20100704 13:18:36< CIA-87> later and slightly farther away and giving Darken Volk less gold on the easier
20100704 13:18:36< CIA-87> difficulties. Also, Dela can now be killed (except on the first turn) but will
20100704 13:18:36< CIA-87> now attempt to set up camp on a location clear of enemies.
20100704 13:30:03< CIA-87> zookeeper * r43887 /branches/1.8/data/campaigns/ (11 files in 3 dirs): Ported r43882-43885 to 1.8.
20100704 13:38:39-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20100704 13:39:52< zookeeper> Ivanovic, at what time would you like to tag?
20100704 13:40:20< Ivanovic> zookeeper: lets say in 3 hours
20100704 13:40:32< zookeeper> i'd like to make cliffs of thoria (HttT) easier but since the WML is such that myth should be kicked in the teeth for it, i need to basically rewrite the whole spawning system to do so...
20100704 13:41:16< zookeeper> ok, 3 hours should be plenty enough
20100704 14:05:42< timotei> Ivanovic: hey
20100704 14:06:26< timotei> wesnoth is using "strict" version growing like: 1.8.2, 1.8.3, ... 1.8.9 , 1.9.0 or it can skip from 1.8.4 to 1.9.0 for example?
20100704 14:07:33< Crab_> timotei: 'stable' wesnoth version are numbered like   1.4.0, 1.4.1, 1.6.0, 1.6.0a, 1.6.1 , 1.6.2, 1.6.3, ... , 1.6.5, 1.8.0, 1.8.1 
20100704 14:07:37< Ivanovic> timotei: stable releases of the 1.8.x series have a strict growing
20100704 14:07:43< Ivanovic> the next trunk release will be 1.9.0
20100704 14:08:26< timotei> so from trunk are only versions like x.x.0, and stables are created from branches?
20100704 14:10:42< Crab_> timotei: trunk version numbers can grow like 1.7.0, 1.7.1, ... , 1.7.15 ; then, when 1.7.x is deemed 'ready', it is branched as next stable version, e.g. 1.8.0
20100704 14:10:51< timotei> oh
20100704 14:12:01< Crab_> after that, 1.8 and trunk ( 1.9 ) lines are independent (through they have a shared history and sometimes fixes get backported from trunk to 1.8 branch)
20100704 14:12:44< timotei> aha
20100704 14:12:52< Crab_> for example, look at the last commit (r43887)
20100704 14:13:05< timotei> yeah, saw the "port"
20100704 14:13:28< Crab_> and note the 'branches' in the url, too
20100704 14:13:35< timotei> right
20100704 14:14:19< Crab_> timotei: take a look at http://svn.gna.org/viewcvs/wesnoth/branches/
20100704 14:14:52< CIA-87> upthorn * r43888 /trunk/src/ (md5.cpp md5.hpp): updated all types in MD5.hpp and MD5.cpp to match with mordante's changes in r43871.
20100704 14:15:11< Crab_> timotei: you can see the `stable release` branches in there, from 1.0 
20100704 14:15:39-!- Appleman1234 [~Appleman1@CPE-60-226-178-176.qld.bigpond.net.au] has quit [Remote host closed the connection]
20100704 14:15:44< timotei> haha, 1.2 is older than 1.0
20100704 14:16:20< Crab_> timotei: it's a 'last touched' date, not creation date.
20100704 14:16:56< timotei> yeahz but ...
20100704 14:17:11< timotei> it's funny :D
20100704 14:17:17< Crab_> yes
20100704 14:20:50< Crab_> Upthorn: it would be a better thing to use mordante's typedefs ( typedef boost::uint32_t uint4; and others) instead of using boost:: directly, in r43888 - otherwise, there's little sense in those typedefs, when half of the code is using them but another half doesn't )
20100704 14:21:15< Upthorn> Crab_: I tried but it didn't seem to recognize them as types
20100704 14:21:59< Upthorn> ... though I guess I didn't try putting md5:: first...
20100704 14:22:22< Crab_> strange - i.e., you have  lines like      ' inline boost::uint32_t MD5::rotate_left  (uint4 x, uint4 n){' in there, so uint4 is a known type :)
20100704 14:26:40< Upthorn> I figured out the problem
20100704 14:26:49< Upthorn> inline uint4 MD5::rotate_left won't work
20100704 14:26:50< Upthorn> but
20100704 14:26:58< Upthorn> inline MD5::uint4 MD5::rotate_left will
20100704 14:27:33< CIA-87> upthorn * r43889 /trunk/src/ (md5.cpp md5.hpp): updated r43888 to use typedefs.
20100704 14:29:04-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Ping timeout: 240 seconds]
20100704 14:47:41< Elvish_Pillager> In WML: I have two variables containing lists of tags (for instance, commands or animations) where the order matters. I want to make a variable that combines the two of them, with all the contents of one coming after the contents of the other. Is there any way to do this?
20100704 14:50:20< silene> lua
20100704 15:00:21< zookeeper> [set_variables] won't do?
20100704 15:00:30< Elvish_Pillager> zookeeper: None of the modes do that
20100704 15:00:39< Elvish_Pillager> unless you can point out a usage I'm missing
20100704 15:00:48< zookeeper> hmmh
20100704 15:01:18< Elvish_Pillager> silene: I'm finding the wiki kind of difficult to use to learn about the lua<->WML interface; is there any other documentation I should be aware of?
20100704 15:02:51< zookeeper> you might want to try [set_variables] ... [insert_tag] name=value variable=onevariable [/insert_tag] [insert_tag] name=value variable=anothervariable [/insert_tag]
20100704 15:02:57< zookeeper> err, actually no
20100704 15:02:59< zookeeper> that's different
20100704 15:02:59< Elvish_Pillager> nope
20100704 15:03:02-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100704 15:03:05< silene> Elvish_Pillager: none that i can think of; in your case, the only functions you need from the interface are get_variable and set_variable
20100704 15:04:11< zookeeper> i guess it might be impossible without lua then
20100704 15:04:37< Elvish_Pillager> silene: Okay; I've finally found where on the wiki it describes the correspondence between Lua variables and WML variables
20100704 15:04:42< zookeeper> (...except i bet you could simply create/use the variables differently)
20100704 15:04:45< Elvish_Pillager> (it's not in the descriptions of get_variable or set_variable...)
20100704 15:05:20< Elvish_Pillager> zookeeper: the precise situation is this: I have an animation and I want to add other [xxx_frame]s to it, but in an [if]-[else] structure.
20100704 15:05:38< Elvish_Pillager> So I have to add the [if] and [else] things directly to the contents of the animation, and they have to be in the right order.
20100704 15:06:06< zookeeper> urgh...
20100704 15:06:42< Elvish_Pillager> I think that using [set_variables] mode=append for each individual [if] and [else] might work, but it's undocumented whether that puts them in the right order.
20100704 15:11:48< Elvish_Pillager> silene: the wiki appears not to document what wesnoth.get_variable() will return if called on an array with multiple elements
20100704 15:12:56< silene> Elvish_Pillager: you mean a sequence of containers? then it behaves the exact same way as plain wml, it returns the one at index 0
20100704 15:13:22< Elvish_Pillager> :(
20100704 15:13:37< silene> but unless i misunderstood, that's not your situation
20100704 15:13:47< Elvish_Pillager> true, it's not my current situation
20100704 15:14:11< Elvish_Pillager> it helps me understand, though, since I was falsely assuming that it had a way to represent the entire (WML) array
20100704 15:14:36< silene> use helper.get_variable_array for that
20100704 15:28:52-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100704 15:36:45< Elvish_Pillager> zookeeper: testing confirms that mode=append can do what I'm looking for. I really should learn this lua stuff though...
20100704 15:50:36-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Read error: No route to host]
20100704 15:50:46-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100704 16:00:36-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Remote host closed the connection]
20100704 16:01:33-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100704 16:01:56< Ivanovic> zookeeper: so what is your current status?
20100704 16:01:57< Ivanovic> are you done?
20100704 16:02:17< Ivanovic> please write some changelog entries for your campaign changes (since those are normally uservisible...)
20100704 16:02:41< zookeeper> Ivanovic, testing my changes atm
20100704 16:03:03< zookeeper> that is, been waiting for HttT to start for the last 5 minutes, haven't gotten to play yet... :p
20100704 16:03:28< Ivanovic> :)
20100704 16:03:45 * Ivanovic is currently trying to setup his system to be able to compile for the pandora
20100704 16:03:50< Ivanovic> compile wesnoth using cmake that is
20100704 16:04:12-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-76.telecom.by] has quit [Remote host closed the connection]
20100704 16:05:23< zookeeper> great, i made a simple WML error, now i have to restart the game
20100704 16:06:00< Ivanovic> outsch
20100704 16:06:18< zookeeper> anyone know how to easily disable the transition bloat?
20100704 16:08:32< zookeeper> good grief, now i can't even simply remove transitions.cfg or any others since then the game chokes on missing macro defs... -.-
20100704 16:10:18< zookeeper> oh well, i guess i can use 1.8 to test it just as well, that's where it'll end up anyway..
20100704 16:14:45< Ivanovic> yeah
20100704 16:14:54-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-76.telecom.by] has joined #wesnoth-dev
20100704 16:24:16-!- wesbot changed the topic of #wesnoth-dev to: 1.8.3 planned for this weekend: stringfreeze for branches/1.8! | 136 bugs, 282 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100704 16:25:08< zookeeper> Ivanovic, ok, will take a little bit longer still...
20100704 16:25:50< Ivanovic> currently working on the openpandora binary anyway...
20100704 16:38:34< zookeeper> i should be ready soon'ish...
20100704 16:40:25-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100704 16:47:13< Ivanovic> and yeah, wesnoth runs *perfectly* on the pandora
20100704 16:47:26< Ivanovic> okay, it takes a while to create the cache (something like 30s to 60s), but it is really fine
20100704 16:47:32< Ivanovic> (only tested 1.8.2 so far)
20100704 16:48:36< timotei> what's pandora?
20100704 16:48:39< timotei> some virtual box?
20100704 16:48:45< timotei> virtual machine*
20100704 16:48:48< Ivanovic> http://www.open-pandora.org/
20100704 16:48:54< Ivanovic> a real geeks toy!
20100704 16:49:05-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100704 16:54:06-!- Johannes13_ [~Johannes@pD95015A4.dip0.t-ipconnect.de] has joined #wesnoth-dev
20100704 16:54:06-!- Johannes13_ [~Johannes@pD95015A4.dip0.t-ipconnect.de] has quit [Changing host]
20100704 16:54:06-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100704 16:54:12< timotei> w00t:O
20100704 16:54:19< timotei> you have such a thing?:O
20100704 16:55:02< Ivanovic> jepp
20100704 16:55:15< Ivanovic> got mine last tuesday, ordered 30th september 2008
20100704 16:56:13-!- timotei [~Timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100704 16:57:22-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 240 seconds]
20100704 16:58:53< Espreon> Two days later and a meme would have been invoked...
20100704 17:02:06< zookeeper> hmmh, i guess i got it working right now
20100704 17:06:05< zookeeper> changelog updates coming in a minute...
20100704 17:06:10< CIA-87> zookeeper * r43890 /trunk/data/campaigns/Heir_To_The_Throne/scenarios/19c_Cliffs_of_Thoria.cfg: 
20100704 17:06:10< CIA-87> A more tolerable spawn rate for the drakes: lvl2 and lvl3 drakes will now take 2
20100704 17:06:10< CIA-87> and 3 turns to spawn, there are no lvl3 spawns on easy and no spawns will occur
20100704 17:06:10< CIA-87> if there's already a drake standing on the spawnpoint. Also rewrote the spawn
20100704 17:06:10< CIA-87> system to be less bloated.
20100704 17:06:39< zookeeper> i just hope i didn't introduce some subtle nasty bugs which i don't have time to test for right now...:P
20100704 17:07:51< CIA-87> zookeeper * r43891 /branches/1.8/data/campaigns/Heir_To_The_Throne/scenarios/19c_Cliffs_of_Thoria.cfg: Ported r43890 to 1.8.
20100704 17:09:12< zookeeper> i think it's probably really easy now, but whatever, at least it's variety
20100704 17:10:35-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Remote host closed the connection]
20100704 17:11:45< zookeeper> also i'll still port my change to DiD to 1.8, but that won't take long at all
20100704 17:14:59< Ivanovic> good
20100704 17:17:00-!- koda [~vittorio@95.236.230.233] has joined #wesnoth-dev
20100704 17:17:14< koda> hello
20100704 17:19:08< zookeeper> Ivanovic, forget the DiD changes, i'd have to sort out map conflicts and i don't have time neither for that or even opening the trunk editor
20100704 17:19:42< Ivanovic> okay
20100704 17:20:33< CIA-87> zookeeper * r43892 /trunk/ (changelog players_changelog): Updated changelogs.
20100704 17:20:34 * zookeeper is really annoyed by trunk
20100704 17:21:58< koda> umh i've been trying to compile wesnoth on the iphone but it prints a lots of compilation errors
20100704 17:22:03< koda> can anyone help me?
20100704 17:22:05< CIA-87> zookeeper * r43893 /branches/1.8/ (changelog players_changelog): Updated changelogs.
20100704 17:23:39-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has joined #wesnoth-dev
20100704 17:25:09< Espreon> koda: If you're trying to compile regular Wesnoth, forget it; if you're trying to compile the iPhone port, ask the porter.
20100704 17:25:45< koda> keypoole?
20100704 17:25:58< zookeeper> if someone else wants to port my latest DiD change to 1.8 by redrawing the map changes in 1.8, then be my guest, the WML should be compatible. also i haven't even tested it yet so maybe it doesn't even work. i have no time to do that.
20100704 17:26:09< zookeeper> so i'm done for now
20100704 17:26:38< Espreon> koda: Yeah, Kyle Poole.
20100704 17:27:51< koda> i met him once, but don't know where to find him
20100704 17:28:03< koda> was hoping that he hung around irc :)
20100704 17:28:21< Espreon> Try PMing him; he'll get an e-mail notification.
20100704 17:30:07< koda> i'll try, thanks
20100704 17:30:35< Espreon> You are welcome.
20100704 17:45:05-!- Blarumyrran [~Blarumyrr@unaffiliated/blarumyrran] has joined #wesnoth-dev
20100704 17:46:29< Ivanovic> AI0867, Crab_, crimson_penguin, esr, Espreon, grzywacz, ilor, loonycyborg, mordante, shadowmaster, silene, Soliton, Upthorn, zookeeper, everyone else who cares: ready for 1.8.3?
20100704 17:47:02< Crab_> 1.8.3 already, time is running fast ;)
20100704 17:47:08< loonycyborg> Ivanovic: Yes.
20100704 17:49:32< crimson_penguin> sure
20100704 17:50:39< Espreon> Sure.
20100704 17:50:48< esr> Ivanovic: OK.
20100704 17:51:08< silene> Crab_: have you seen report https://gna.org/bugs/?16223 ? behavior of ai leaders is broken in 1.8
20100704 17:51:29-!- e_s-iOS [~esios@pool-108-2-88-55.phlapa.east.verizon.net] has quit [Remote host closed the connection]
20100704 17:51:44< Crab_> silene: no, haven't seen it yet.
20100704 17:54:36< CIA-87> ivanovic * r43894 /branches/1.8/po/ (231 files in 23 dirs): 
20100704 17:54:36< CIA-87> pot-update (reference update only, no new/changed strings, preparing for 1.8.3)
20100704 17:54:36< CIA-87> regenerated doc files (no changes here)
20100704 17:56:39< Crab_> silene: thanks. I'll fix, but not for 1.8.3
20100704 17:57:38< CIA-87> ivanovic * r43895 /trunk/po/ (wesnoth-ei/vi.po wesnoth-l/vi.po): updated Vietnamese translation
20100704 17:57:42< CIA-87> ivanovic * r43896 /branches/1.8/po/ (wesnoth-ei/vi.po wesnoth-l/vi.po): updated Vietnamese translation
20100704 18:07:55< CIA-87> ivanovic * r43897 /branches/1.8/ (changelog configure.ac players_changelog src/wesconfig.h): bump version to 1.8.3
20100704 18:25:59< CIA-87> ivanovic * r43898 /tags/1.8.3/: tagged 1.8.3, using r43897
20100704 18:27:00-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100704 18:27:31< CIA-87> ivanovic * r43899 /branches/1.8/ (changelog configure.ac players_changelog src/wesconfig.h): post release version bump to 1.8.3+svn
20100704 18:27:33-!- Ivanovic changed the topic of #wesnoth-dev to: released 1.8.3, announcing "soon" | 136 bugs, 282 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100704 18:27:43-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100704 18:40:54-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 252 seconds]
20100704 18:42:15-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100704 18:42:50-!- EVILEST [~Art@S01060021297e1255.vc.shawcable.net] has joined #wesnoth-dev
20100704 18:45:35-!- EVILEST [~Art@S01060021297e1255.vc.shawcable.net] has left #wesnoth-dev []
20100704 18:56:07< mordante> Crab_, I only updated the bad typedef, assumed the code used the properly (compiles for me)
20100704 18:56:56< mordante> Crab_, Upthorn IIRC asio depends on date time
20100704 18:56:56< Crab_> mordante: well, not all parts of the code used the typedef. anyway, it was fixed (not in the best way)
20100704 18:57:12< mordante> that would also make sense since we added boost asio
20100704 18:57:13< Crab_> mordante: yes, that what I thought, too..
20100704 19:00:18< mordante> Upthorn, regarding r43888 actually you should have used uint4 (else the typedef makes no sense)
20100704 19:01:20< mordante> Upthorn, nevermind I see you did it in your second commit
20100704 19:03:47< mordante> Ivanovic, I'm ready for 1.8.3
20100704 19:07:02< Ivanovic> Rhonda, crimson_penguin, loonycyborg: 1.8.3 is done
20100704 19:07:44< mordante> Ivanovic, nice to hear Wesnoth runs well on the Pandora :-)
20100704 19:08:44< Ivanovic> uploading the pandora package for 1.8.3 to sf.net, too
20100704 19:08:54< mordante> :-)
20100704 19:09:00< crimson_penguin> updating svn
20100704 19:15:11-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100704 19:16:05-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100704 19:16:40< Elvish_Pillager> The wiki states that you can have multiply [if] blocks in one animation. I have tried this, and it appears buggy. Can anyone confirm or deny that the engine is supposed to be able to handle this?
20100704 19:17:01< Elvish_Pillager> *multiple
20100704 19:17:50< Elvish_Pillager> Specifically, the choice of all but the last [if]-[else] seem to be completely at random every time the animation happens.
20100704 19:22:38-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100704 19:22:52< zookeeper> Elvish_Pillager, there's bugs in it, yes
20100704 19:23:42< zookeeper> especially with multiple [if]s, i think
20100704 19:24:34< zookeeper> or multiple [else]s...i'm not sure what it is exactly, but something like that
20100704 19:25:18< silene> Elvish_Pillager: there is no obvious bugs in the code though; as long as an [else] is right after an [if], it is supposed to work
20100704 19:27:33< zookeeper> the last time i checked, the thunderer doesn't always/usually play the thunderstick miss animations
20100704 19:28:01< zookeeper> and it has a pretty simple if/else structure
20100704 19:29:10< zookeeper> oh, looks like it's been fixed
20100704 19:29:14< zookeeper> nevermind then
20100704 19:32:07< zookeeper> wtf...part of my impression of it not working was because there was no miss sound, but apparently that's just because someone has given it a totally inaudibly quiet miss sound -.-
20100704 19:33:07-!- Mythological [Mythologic@77.29.0.79] has joined #wesnoth-dev
20100704 19:35:09< zookeeper> hello Mythological
20100704 19:35:43-!- Shakey [HydraIRC@c-71-201-89-187.hsd1.il.comcast.net] has joined #wesnoth-dev
20100704 19:35:55< zookeeper> Mythological, i just (probably) made cliffs of thoria ridiculously easy in 1.8.3 a few hours ago ;)
20100704 19:39:49-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100704 19:42:41< Elvish_Pillager> silene: The wiki implies that you can have [if] [else] [else] [else] for more different possibilities; is that not true?
20100704 19:44:05< silene> Elvish_Pillager: yes, it is, i just meant to say that, contrarily to event wml, the order matters
20100704 19:44:12-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has quit [Client Quit]
20100704 19:46:28< Elvish_Pillager> perhaps it's a problem that my different [if] blocks contain different types of frames (e.g. [frame] vs. [missile_frame])?
20100704 19:47:16< silene> no, the code doesn't care
20100704 19:47:43< silene> (just so that there is no misunderstanding; i said there was no obvious bug, i didn't say there was no bug)
20100704 19:47:50< Elvish_Pillager> right
20100704 19:48:26< Elvish_Pillager> I haven't tried much of a variety of stuff, but I haven't found any situation where having multiple [if]s works
20100704 19:48:34< Elvish_Pillager> and I don't see any in mainline
20100704 19:48:46< Elvish_Pillager> (although there could be something I missed)
20100704 19:49:55< zookeeper> the addition of poisoning and slowing sounds usually results in two [if]s
20100704 19:50:10< zookeeper> also the rocklobbing anim might have two, not sure
20100704 19:51:35-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100704 19:51:40< gabba> bonjour
20100704 19:51:49< mordante> hi gabba 
20100704 19:51:54< gabba> hi mordante
20100704 20:05:19-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100704 20:05:35< billynux> hi mordante 
20100704 20:05:42-!- gabba [~gabba@wesnoth/developer/gabba] has quit [Quit: Leaving.]
20100704 20:19:07-!- timotei [~Timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100704 20:21:26< loonycyborg> Ivanovic: 1.8.3 windows installer is uploaded.
20100704 20:21:36< Ivanovic> loonycyborg: cool
20100704 20:22:12< Elvish_Pillager> zookeeper: the good news is that I didn't need the [if]-[else], and now my wolf-riding wizards' graphics are cleaner than ever
20100704 20:23:11-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100704 20:23:41-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100704 20:24:12< zookeeper> Elvish_Pillager, don't say you made a wolf-riding wizard by creating an animation which just combines the wolf and mage sprites using sub-animations? :x
20100704 20:24:31< Elvish_Pillager> Nope, I did it a while ago
20100704 20:24:36< Elvish_Pillager> It actually looks pretty nice
20100704 20:24:42< zookeeper> right
20100704 20:24:54< Elvish_Pillager> did it for water serpents, gryphons, dragons, and mudcrawlers too
20100704 20:29:59< Elvish_Pillager> the new way I'm doing it is using the wizard (Red/Arch/Great Mage or Ancient/Lich actually) as the base [frame] and cutting the monster image into different [xxx_frame]s so that I can choose which parts obscure the wizard and which don't
20100704 20:34:21< zookeeper> uh, ok
20100704 20:35:16< Elvish_Pillager> Not interested? :p
20100704 20:38:03< zookeeper> sounds a bit too hacky for my taste ;)
20100704 20:38:55< Elvish_Pillager> pfft, I can deal with a little hacky if the payoff is gryphon riding wizards
20100704 20:40:41-!- Mythological [Mythologic@77.29.0.79] has quit [Ping timeout: 240 seconds]
20100704 20:40:58-!- Mythological [Mythologic@77.29.6.89] has joined #wesnoth-dev
20100704 20:44:55< Elvish_Pillager> w00t, segfault!
20100704 20:47:40-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Read error: No route to host]
20100704 20:50:39-!- ilor [~user@wesnoth/developer/ilor] has quit [Ping timeout: 245 seconds]
20100704 20:57:26-!- ilor [~user@wesnoth/developer/ilor] has joined #wesnoth-dev
20100704 21:06:05-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100704 21:08:50-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20100704 21:14:11< timotei> hm, I've done some 5 types of modification in my files. in the commiting "gentlemen rules" was stated to split your commits between smaller ones
20100704 21:14:26< timotei> is problem if I commit in a single one, since are dependent between them
20100704 21:15:00< billynux> timotei, do you have a link to these "commiting rules" ?
20100704 21:15:13< timotei> hey billynux 
20100704 21:15:15< timotei> one sec
20100704 21:15:41< timotei> no huge commit (hard to review), so when possible maybe split it in working part with info about where you are going 
20100704 21:15:45< timotei> sorry: http://wiki.wesnoth.org/DeveloperGuide
20100704 21:16:00< billynux> ok
20100704 21:16:31< timotei> also this is nice too: http://wiki.wesnoth.org/CodingStandards
20100704 21:16:41< billynux> timotei, I'm having trouble with the current version. "WML exception: User message: Window not defined." in settings.cpp:283
20100704 21:16:47< billynux> was that you? :)
20100704 21:16:56< timotei> wow, happly not :D
20100704 21:17:03< timotei> I'm not working (so much) on the C++ side
20100704 21:17:28< timotei> one second, we can check who's "guilty"
20100704 21:17:42< timotei> settings.cpp it's in: src/settings.cpp right?
20100704 21:17:50< billynux> I'm not into finding the "guilty" one, but the solution
20100704 21:18:05< billynux> last change was by fendrin 
20100704 21:18:11< timotei> well maybe who modified that could know the problem:-/
20100704 21:18:30< timotei> wow, 2 months back there
20100704 21:18:33< billynux> fendrin, any idea why I'm having that problem ^ ?
20100704 21:18:50< Crab_> billynux: take a look, which window is not defined.
20100704 21:19:41< billynux> well, src/settings.cpp doesn't have 283 lines. Ok Crab_ 
20100704 21:19:54< Crab_> it's src/gui/widgets/settings.cpp
20100704 21:19:59< billynux> src/gui/widgets/settings.cpp:283 in function 'read'.
20100704 21:20:00< billynux> yep
20100704 21:20:32< Crab_> as you see, there's something in registered_window_types which is not in window_types
20100704 21:20:48< billynux> how do I check "which window is not defined" Crab_ ? It's when trying to render the initial screen
20100704 21:21:13< Crab_> billynux: via a debugger 
20100704 21:21:26< billynux> k
20100704 21:22:01< Crab_> billynux: or you can try to guess...
20100704 21:22:38< silene> billynux: are you running your binary with the data from the same revision? if not, that's not an uncommon error
20100704 21:22:49-!- ElCaballero [~saul@201.243.19.27] has joined #wesnoth-dev
20100704 21:22:55-!- ilor [~user@wesnoth/developer/ilor] has quit [Ping timeout: 260 seconds]
20100704 21:23:56< billynux> silene, maybe not, I compiled and run previous versions just fine
20100704 21:24:13< billynux> do I have to delete something so it would generate new data?
20100704 21:24:52< silene> no, you don't have to delete anything; you just have to make that your data/ directory is not more recent (or older) than your wesnoth binary
20100704 21:24:57< Crab_> billynux: for example, data_manage window was recently added to the source code, it's in /wesnoth/data/gui/default/window/data_manage.cfg and /wesnoth/src and /wesnoth/src/gui/dialogs/data_manage.cpp
20100704 21:27:03< billynux> ok, but where did I go wrong? Should I have deleted CMake's cache for recompilation?
20100704 21:27:32< billynux> (I did svn up and make in my CMake's build dir)
20100704 21:28:01< silene> billynux: how do you run wesnoth?
20100704 21:28:27< silene> (are you passing the uptodate data/ location as an argument?)
20100704 21:28:35< billynux> right now, its a debug version, so gdb ./wesnoth, and then r
20100704 21:28:44< billynux> silene, no
20100704 21:29:10-!- ilor [~user@wesnoth/developer/ilor] has joined #wesnoth-dev
20100704 21:29:26< billynux> -datadir <dir> I'm guessing
20100704 21:30:33< Crab_> billynux: I'd use a debugger or just `add code&recompile` to output the offending *itor to log - it's a string id of the offending window.
20100704 21:32:09< Crab_> billynux: then you would know what files you must check. for example, if window data_manage would be absent, you'd need to check the path to data dir in the log, and the data/gui/default/window/data_manage.cfg file in it
20100704 21:33:22-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100704 21:33:25< billynux> mmm... checking... I'm in gdb, just before the error occurs
20100704 21:33:34< Crab_> billynux: bonus points if you improve the error message, for the future :)
20100704 21:33:42< billynux> :P
20100704 21:34:36< billynux>  (gdb) p *itor
20100704 21:34:36< billynux> $1 = (const std::basic_string  / . . .  /   "wml_message_left"
20100704 21:34:50< Crab_> billynux: it basically does 'foreach string window_id which is registered  in the source, check if we have its definition from the data directory, if not , throw exception'
20100704 21:36:03< Crab_> billynux: it's the one which fails the VALIDATE ?
20100704 21:36:10< billynux> -datadir ~/code/wesnoth/data gives me: "error filesystem: Trying to open file with empty name.", "error config: Could not open file", "could not initialize fonts"
20100704 21:36:19< billynux> Crab_, let me check
20100704 21:36:28< Crab_> billynux: or it's the first which is checked ?
20100704 21:36:58< silene> billynux: don't pass data, pass its parent
20100704 21:37:10< billynux> duh :P, sorry
20100704 21:37:25< billynux> yep, that work
20100704 21:37:28< billynux> *works
20100704 21:37:43< billynux> Crab_, is there any point in figuring out which one fails now?
20100704 21:38:43< Crab_> billynux: it works with the path to data directory explicitly set ?
20100704 21:38:49< billynux> and, wml_message_left was the first, not the faulty one
20100704 21:40:03< billynux> Crab_, the problem was "data_manage", so yes, just a problem between my data versions
20100704 21:40:23< Crab_> billynux: I'd say that the error message needs to be improved.
20100704 21:40:25< billynux> Crab_, yes... sorry, didn't know about this issue
20100704 21:40:32-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100704 21:40:55< Crab_> billynux: it should at least log the offending window name, and it should tell that this might be caused by data/src mismatch
20100704 21:41:04< billynux> something like: "Maybe you recompiled and forgot to use the updated data directory." :)
20100704 21:41:27< Crab_> billynux: it would be a great thing if you fix it to be both informative and user-friendly, while you're at it. )
20100704 21:41:42< billynux> change the VALIDATE with a custom str?
20100704 21:42:35< Crab_> check other uses of VALIDATE, there should be a suitable example
20100704 21:43:30< Crab_> billynux: for example, ./map.cpp +182
20100704 21:43:43< Ivanovic> mordante: in case your are interested: howto compile and package wesnoth for the openpandora: http://forums.wesnoth.org/viewtopic.php?f=5&t=30640
20100704 21:45:29< Elvish_Pillager> how stupid of me: segfaults are probably the kind of things I should expect when cropping an image starting from beyond the right edge
20100704 21:45:33< Crab_> billynux: also note that gdb has 'catch throw' command to make it stop when exception is thrown, this is handy for debugging VALIDATE calls
20100704 21:46:25< billynux> Crab_, was thinking something like -> const std::string error_msg("Window not defined: '" + *itor + "'. Perhaps a mismatch between data and source versions.");
20100704 21:46:55< Crab_> billynux: better to do it in a more translator-friendly way :)
20100704 21:47:23< Crab_> billynux: i.e. use variable substitution as in ./map.cpp +182 instead of string concatenation
20100704 21:47:43< Crab_> billynux: but yes, the message contents look fine.
20100704 21:48:18< Crab_> you can even state that 'window not defined in WML' or 'window not defined in data' 
20100704 21:48:41< Crab_> as this particular error is 'window is in source but not in data'
20100704 21:49:31< billynux> ok, I'm not familiar with the translation implementation. I'm looking into it.
20100704 21:50:43< silene> Crab_: please no translation for that; no need to give more work to translators that will never happen for the end user
20100704 21:50:55< silene> +for an error
20100704 21:51:57< Crab_> silene: note: it was translatable before
20100704 21:52:32< silene> even better, it means the amount of work actually diminished
20100704 21:52:36< Crab_> ok
20100704 21:53:46< timotei> Ivanovic: some pictures with it?:D
20100704 21:53:57< timotei> with it, meaning: running on your pandora
20100704 21:54:00< billynux> Maybe this is a newbie question for Wesnoth developer, but what does _( ... ) mean? Is there a class defined as "_" ?
20100704 21:54:04< Ivanovic> timotei: the only camera i got is in my mobile phone
20100704 21:54:11< timotei> :)
20100704 21:54:13< Ivanovic> can't get really decent pics with that one...
20100704 21:54:21< silene> billynux: no, it's a function (actually a macro to the gettext function)
20100704 21:54:53< billynux> isn't that an insane function name? :)
20100704 21:55:06< silene> billynux: note that it doesn't have anything to do with wesnoth, you will find it in any software a bit serious about internationalization
20100704 21:55:06< billynux> Ok... User message: Window not defined: 'data_manage'. Perhaps a mismatch between data and source versions.
20100704 21:55:47< Crab_> billynux: looks good. 'Window not defined in WML:' would be slightly better :)
20100704 21:55:48< billynux> Should I commit that? it's still VALIDATE( ... , _( error_msg.c_str() ); <- with _( there
20100704 21:55:55< silene> no!
20100704 21:56:06< Crab_> no, if no translation, then there's no need for _()
20100704 21:56:07< silene> remove the _(), it only works with literal strings
20100704 21:56:10< billynux> good
20100704 21:57:18< billynux> Should I add "try -datadir <wesnoth-code-dir>." there?
20100704 22:01:25< mordante> hi billynux seems we finally manage to meet on irc :-)
20100704 22:01:47< billynux> yeah! :)
20100704 22:03:25-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Quit: I hope y'all have a nice day ;)]
20100704 22:04:07< billynux> Crab_, silene, check -v
20100704 22:04:08< CIA-87> billynux * r43900 /trunk/src/gui/widgets/settings.cpp: Improved error message when using different source and data versions.
20100704 22:05:24< billynux> mordante, so... as you know I'm working on logging to the main servers using ana. I get to the 'get id' part of the handshake ATM
20100704 22:05:32< mordante> great Ivanovic 
20100704 22:06:11< mordante> Ivanovic, how generic is the pandora toolchain file? if generic maybe commit it 
20100704 22:06:19< billynux> my PC takes 5 minutes to -link- wesnoth :(
20100704 22:06:31< Crab_> billynux: ok
20100704 22:06:53< shadowmaster> billynux: how much RAM does it have? are you making debug builds or release builds?
20100704 22:07:12< billynux> shadowmaster, 1GB, and a debug build
20100704 22:07:25< Ivanovic> mordante: does not make much sense
20100704 22:07:32< mordante> too bad
20100704 22:07:38< shadowmaster> you need at least 4 GB for making debug builds without running out of memory, billynux :P
20100704 22:07:41< Ivanovic> mordante: you have to directly give the path to the file anyway
20100704 22:08:00< shadowmaster> (which is why I recently added 2 GB)
20100704 22:08:02< Ivanovic> beside this you have to alter it to point to the paths where your toolchain is installed
20100704 22:08:13< mordante> I know but writing toolchain files is tricky for some people
20100704 22:08:26< Ivanovic> mordante: this is a 1:1 copy from the official cmake wiki
20100704 22:08:36< mordante> ok
20100704 22:08:37< billynux> shadowmaster, :P... I'll upgrade my PC with GSoC money soon. Pending mordante's positive review :) And my laptop works just fine (4GB, core2duo)
20100704 22:08:45< Ivanovic> just altered the three paths to point to the correct gcc, g++ and root of the toolkit
20100704 22:08:56< mordante> billynux, positive reviews are in your own hand ;-P
20100704 22:09:13< timotei> ah, I miss Visual studio's conditional breakpoints :-<
20100704 22:09:27< mordante> billynux, I had a look at the code not much comment only the heavy use of inlined member functions is not really C++ like
20100704 22:09:52< mordante> and best split in a (private) header and then define the implementation seperately
20100704 22:09:53< shadowmaster> timotei: you can do that with gdb/kdevelop IIRC
20100704 22:09:59< shadowmaster> unless I'm missing the "conditional" part
20100704 22:10:01< timotei> shadowmaster: but not with eclipse :P
20100704 22:10:22< billynux> mordante, you mean network_ana.cpp right?
20100704 22:10:24< timotei> welll, when placed a breakpoint, right click on the breakpoint and put a "local" condition when that breakpoint should trigger
20100704 22:10:28< mordante> C++ normally inlines member functions defined in a class, so you should only do that if they're really short and 10+ lines is too much
20100704 22:10:30< mordante> yes
20100704 22:10:41< timotei> basically can be written in normal code, but yeahz... when you need to DEBUG it ...
20100704 22:11:02< mordante> and personally I prefer files with one class and split the code in multiple files (not that you need to do that per se)
20100704 22:11:18< billynux> mordante, I agree. Should I define them below? (but still in the same file)
20100704 22:11:32< mordante> but the files is already over 1000 lines so slowly starts to become hard to navigate
20100704 22:11:52< mordante> either that or a (private) header
20100704 22:12:08< billynux> mordante, true... but I will likely remove a lot of code from there.
20100704 22:12:31< mordante> from the functions or the file?
20100704 22:12:34< billynux> mordante, that does sound better. ana_manager.hpp and ana_manager.cpp?
20100704 22:12:42-!- Blueblaze [~Blueblaze@adsl-99-148-247-235.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100704 22:12:51< mordante> to me it does
20100704 22:13:17< billynux> I meant from the file (network_ana.cpp) I think there are some things I won't end up using
20100704 22:13:36< mordante> ok 
20100704 22:13:57< billynux> mordante, and everybody else: After a handshake to a wesnoth server. What is the scope of the received id? 
20100704 22:14:25< billynux> meaning: will other code use *this* id as the id of the component, and then issue something like send( *this-id* ... ?
20100704 22:14:58< billynux> semantics: It's the ID *you* have in the wesnoth server right?
20100704 22:15:45-!- dtiger [~dtiger@dynamic-vpdn-93-125-17-76.telecom.by] has quit [Remote host closed the connection]
20100704 22:16:58< mordante> in the current implementation?
20100704 22:18:03< billynux> yes
20100704 22:18:57< billynux> I'm asking to know what to do with the ID, during the handshake, you send a 0 for a new connection or a handle for a reconnection
20100704 22:19:17< billynux> I'm guessing this *received* id is your handle
20100704 22:19:38< mordante> brb
20100704 22:19:54< timotei> fendrin: hey
20100704 22:20:27< timotei> fendrin: I've finished the part of "importing" the user add-ons dir, and today I'll go ahead finishing the wizards
20100704 22:20:46< timotei> ok, here it comes commit flood
20100704 22:20:47< timotei> :))
20100704 22:21:34< timotei> damn, forgot to update first:-S
20100704 22:21:37< CIA-87> timotei * r43901 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/wizards/CampaignNewWizard.java: eclipse plugin: small refactor on array initialization
20100704 22:21:58< timotei> oh pfew, thanks god git is intelligent enough :))
20100704 22:23:57< CIA-87> timotei * r43902 /trunk/utils/java/eclipse_plugin/plugin.xml: eclipse plugin: add plugin preferences menu in Main Menu
20100704 22:24:10< CIA-87> timotei * r43903 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/builder/WesnothProjectBuilder.java: eclipse plugin: revert a comment
20100704 22:24:23< CIA-87> timotei * r43904 /trunk/utils/java/eclipse_plugin/ (4 files in 4 dirs): eclipse plugin: add the 'setup workspace' menu entry
20100704 22:24:34< CIA-87> timotei * r43905 /trunk/utils/java/eclipse_plugin/plugin.xml: eclipse plugin: give better names for the builder and nature
20100704 22:24:51< CIA-87> timotei * r43906 /trunk/utils/java/eclipse_plugin/ (3 files in 2 dirs): eclipse plugin: initial files for era and multiplayer templates
20100704 22:25:03< CIA-87> timotei * r43907 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/builder/WesnothProjectBuilder.java: eclipse plugin: use a monitor to inform the user about builder's progress
20100704 22:25:18< CIA-87> timotei * r43908 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/WorkspaceUtils.java: eclipse plugin: add the .ignore file for skipping existing projects in workspace
20100704 22:25:28< CIA-87> timotei * r43909 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/ResourceUtils.java: eclipse plugin: return null if there was an exception
20100704 22:25:40< CIA-87> timotei * r43910 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/WorkspaceUtils.java: eclipse plugin: use the location full path rather than directory name
20100704 22:25:51< CIA-87> timotei * r43911 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/WorkspaceUtils.java: eclipse plugin: the "user addons" project doesn't need to be converted
20100704 22:26:02< CIA-87> timotei * r43912 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/ResourceUtils.java: 
20100704 22:26:03< CIA-87> eclipse plugin: check for file existance before using it
20100704 22:26:03< CIA-87> add the 'overwrite' flag for overwriting existing files if any
20100704 22:26:13< CIA-87> timotei * r43913 /trunk/utils/java/eclipse_plugin/templates/build.xml: eclipse plugin: rename a replaceable parameter in the build.xml template
20100704 22:26:26< CIA-87> timotei * r43914 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/wizards/TemplateProvider.java: eclipse plugin: temporary fix for xml-type files comments on empty parameters
20100704 22:26:37< CIA-87> timotei * r43915 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/ (utils/WorkspaceUtils.java wizards/CampaignNewWizard.java): eclipse plugin: don't create an auxiliary directory for 'User addons' project
20100704 22:29:29< billynux> I usually try to commit single purpose changes, but it may span several files
20100704 22:30:43< timotei> billynux: well, git just got me a : <timotei> FATAL ERROR: Network error: Connection timed out
20100704 22:30:43< timotei> <timotei> Network connection closed unexpectedly: Connection closed unexpectedly at C:\Program Files\Git/libexec/git-core/git-svn line 2290
20100704 22:30:45< timotei> freaking error
20100704 22:31:26< billynux> :(
20100704 22:34:41< timotei> well I'm lucky it crashed ONLY at the last commit :))
20100704 22:35:26< CIA-87> timotei * r43916 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/ (3 files in 2 dirs): eclipse plugin: add the ignore capability in the wesnoth project builder
20100704 22:43:33< mordante> timotei, git is quite smart :-)
20100704 22:43:46< timotei> mordante: yeah, lucky me :D
20100704 22:44:08< mordante> just don't ask shadowmaster about how smart git is ;-)
20100704 22:44:09 * mordante hides
20100704 22:45:08< mordante> billynux, not 100% sure but I assume that's the case, at least it would make sense
20100704 22:45:24< billynux> ok
20100704 22:45:28< timotei> ok guys, I'm out :D
20100704 22:45:41< mordante> billynux, you have any questions at the moment or things you want to discuss?
20100704 22:45:44< mordante> bye timotei 
20100704 22:45:47< timotei> good night mordante , billynux , Crab_ , Ivanovic 
20100704 22:45:55< Ivanovic> n8 timotei
20100704 22:46:15< billynux> mordante, nothing ATM, I'm separating into these 2 new files
20100704 22:46:25< mordante> ok great
20100704 22:46:29-!- timotei [~Timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100704 22:46:31< billynux> so, network_ana.cpp will include "network_manager_ana.hpp"
20100704 22:46:53< mordante> billynux, I hope to be more online the next weeks, but still best send an email if you have an urgent question
20100704 22:46:59< billynux> which will then link to the object file compiled from network_manager_ana.cpp 
20100704 22:47:46< billynux> mordante, ...or, I could define the methods from network_manager_ana.hpp in network_ana.cpp directly, so I won't need to change CMake/SCons/autotools
20100704 22:48:00< mordante> kind of, but if you add it to the project files and use `include' the build system will figure it out
20100704 22:48:20< billynux> yes... not much work
20100704 22:48:31< billynux> so... Add 2 new files or 1?
20100704 22:48:42< billynux> I prefer 2, better organization
20100704 22:48:46< Ivanovic> yeah, build systems can be quite smart these days
20100704 22:49:00< fendrin> damn, missed timotei by a few minutes.
20100704 22:49:00< billynux> and we still use autotools :(
20100704 22:49:22< Ivanovic> (i first tried to get wesnoth compile with cmake altering every single found path, hell, searching for some secs gave me the neat solution...)
20100704 22:49:26< Ivanovic> billynux: and?
20100704 22:49:28< mordante> billynux, 2 and just change the build system files they are meant to do this part of determining what to link when
20100704 22:49:47< Ivanovic> billynux: only silene does use that one and honestly, if it finally dies i don't care too much
20100704 22:49:49< Ivanovic> ;)
20100704 22:50:29< billynux> :)
20100704 22:50:44< mordante> adding files to autotools (or scons and cmake) isn't hard only if libraries change you need to know more about the system
20100704 22:51:07< silene> Ivanovic: yes, and debian, and ubuntu, and... ;-)
20100704 22:51:49< mordante> silene, they can change as well, but as long as one developer wants to maintain it we should keep it
20100704 22:52:04< Ivanovic> silene: rhonda can easily switch
20100704 22:52:47< silene> Ivanovic: easily? when was the last time you packaged something? since when did it become "easy" to start from scratch with a new build system?
20100704 22:53:08< Ivanovic> last time i packaged something: 4h ago, the pandora package
20100704 22:53:14< Ivanovic> ;)
20100704 22:53:39< Ivanovic> silene: the neat thing is that we basically got 1:1 replacements of the existing commands
20100704 22:56:06-!- ancestral [~ancestral@166.137.139.94] has joined #wesnoth-dev
20100704 23:00:12< billynux> mordante, these next 2 weeks I should be doing GSoC only stuff :) (winter vacations here)
20100704 23:01:55< mordante> winter... full summer over here ;-)
20100704 23:02:16< mordante> and hot (at least for this region)
20100704 23:02:40< mordante> but nice that you'll be working full time on gsoc :-)
20100704 23:03:17-!- ancestral [~ancestral@166.137.139.94] has quit [Quit: Colloquy for iPhone - http://colloquy.mobi]
20100704 23:03:33< billynux> Oh... mock me not! At least winters are tolerable here (not in Montreal though, I lived there as well).
20100704 23:04:13< loonycyborg> I wonder when people living on equator have winder vacations :P
20100704 23:04:16< Ivanovic> yesterday we had some 36/37°C over here
20100704 23:04:18< loonycyborg> *winter
20100704 23:04:28< Ivanovic> (perfect timing, went well together with the german victory...)
20100704 23:04:59< billynux> Yes, a friend of mine sent me a funny email from the university at Eindhoven telling people necessary precautions for hot weather like 30C...!!
20100704 23:04:59 * Espreon wants Spain to win for a lulzy reason... if Spain is still in...
20100704 23:05:21< Ivanovic> Espreon: spain will lose in the semi finals
20100704 23:05:23< Ivanovic> against germany
20100704 23:05:32 * billynux Congrats Mark and other netherlanders for their perfomance so far... and is saddened by Argentina/Germany :(
20100704 23:06:07< Ivanovic> billynux: no reason to be sad, you saw the world champion 2010 playing
20100704 23:06:22< billynux> Ivanovic, tread carefully :P
20100704 23:06:26< mordante> billynux, your friend underestimated, Einhoven was well over 30 degrees
20100704 23:06:27< Espreon> Ivanovic: No, Spain has to win... although I wish that I could have Germany win instead.
20100704 23:06:39< Ivanovic> Espreon: spain musn't win
20100704 23:07:02< Ivanovic> it would completely destroy the euro if the players got paid such a huge amount of money for winning the cup
20100704 23:07:08< mordante> billynux, I actually don't care about football, last weekend some Germans told me the Netherlands won ;-) 
20100704 23:07:09< Ivanovic> >half a million *each*
20100704 23:07:20< Ivanovic> (yeah, spain is basically bankrupt)
20100704 23:07:37< Espreon> Is Chile still in?
20100704 23:07:37< Ivanovic> mordante: the netherlands will lose in the finals against germany
20100704 23:07:43< billynux> Espreon, nope
20100704 23:07:46 * Espreon doesn't pay attention to nonsense
20100704 23:07:46< Ivanovic> Espreon: no, they lost against brasil
20100704 23:07:48< Espreon> Good.
20100704 23:08:11< mordante> Ivanovic, I hope not, simply hope we won't enter the finals ;-)
20100704 23:08:13< billynux> Good? Are you chilean too? shadowmaster doesn't care much about football too
20100704 23:08:27< Ivanovic> mordante: but, you know, against uruguay?
20100704 23:08:29< Espreon> billynux: Uh.... NO!
20100704 23:08:37< Ivanovic> anyway, time for a shower
20100704 23:08:39< Espreon> Soy estadounidense.
20100704 23:08:51< mordante> Ivanovic, didn't know whe had to play against Uruguay ;-)
20100704 23:08:58< billynux> :) yeh, enough nonsense, back to work :)
20100704 23:09:04< mordante> good idea
20100704 23:09:20< mordante> the train on Friday was annoying enough everybody listening to football
20100704 23:09:23< shadowmaster> someone mentioned football soccer?
20100704 23:09:43< mordante> unfortunately yes
20100704 23:10:31< billynux> the annoying part to me is: 1- No sports journalist ever talks about sports 2- Everybody forgets about any other issues as long as football is going fine here
20100704 23:10:45< Espreon> billynux: But seriously, where did you get the idea that I was Chilean?
20100704 23:11:11< billynux> Espreon> Is Chile still in? .... no .... Espreon> Good.
20100704 23:11:40< Espreon> I see...
20100704 23:15:52< mordante> billynux, regarding r43900 the message should be translatable we want to try to give the user sane information
20100704 23:16:09< mordante> and casting to a c_str() looks really weird
20100704 23:16:22< billynux> mordante, we discussed this with silene and Crab_ 
20100704 23:16:49< mordante> billynux, I know, but I added VALIDATE so we could give users better translated messages
20100704 23:16:49< billynux> silene's point being: no need to add extra burden in making error messages translatable
20100704 23:17:20< mordante> all VALIDATE messages are translatable and IMO that improves the quality of Wesnoth
20100704 23:17:38< mordante> not all UMC creators are fluent in English
20100704 23:18:25< silene> mordante: umc creators that compile wesnoth are; this particular error can only occur due to a compilation/installation mistake, the end user will never see it
20100704 23:18:26< mordante> also if you look at the function signature it specifically mentions the message should be translatable
20100704 23:19:14< silene> mordante: i'm against cluttering translations with messages that aren't useful for the end user
20100704 23:19:15< mordante> silene, no not all umc creators are fluent
20100704 23:19:38< silene> mordante: you are missing the point, they have to compile wesnoth themselves to experience this message
20100704 23:20:08< mordante> not per se they may also define their own theme stuff
20100704 23:20:31< mordante> but if you feel about it like that it shouldn't be a VALIDATE message
20100704 23:20:56< mordante> VALIDATE is explicitly added to be translatable
20100704 23:21:51< silene> why? it doesn't take a translatable string, it takes a translated string
20100704 23:23:02< silene> if you want it to take a translatable string, just do so, and it will actually reduce the size of wesnoth binary
20100704 23:23:58< mordante> since the macro signature says it wants a translatable string 
20100704 23:25:13-!- un214 [~quassel@adsl-75-45-4-19.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100704 23:25:29< silene> mordante: just because the comment says it wants a translatable string doesn't mean it actually; the code expects a *translated* string!
20100704 23:25:39< silene> not a translatable one
20100704 23:25:45< un214> lol
20100704 23:26:49< billynux> silene, your point is: if it would require a translatable string it would try to translate it before showing it? (or require some base class of translatable strings)
20100704 23:27:57-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100704 23:28:03< silene> billynux: right; translated string: VALIDATE(..., _("something")); translatable string: VALIDATE(..., N_("something"); there is only one character difference, but it behaves completely differently
20100704 23:28:15< boucman> hey all
20100704 23:28:19< Crab_> hi, boucman
20100704 23:28:49< Ivanovic> hi boucman
20100704 23:28:54-!- Blarumyrran [~Blarumyrr@unaffiliated/blarumyrran] has quit [Read error: Connection reset by peer]
20100704 23:29:02< Ivanovic> boucman: speed for creating the cache really needs to be improved somehow
20100704 23:29:12< boucman> yeah yeah, I know...
20100704 23:29:23< Ivanovic> boucman: even on my rather fast system it feels like ages till wesnoth is started if there was some change in the cfg files
20100704 23:29:41< Ivanovic> (no, i have not tried trunk on the pandora so far and honestly i don't plan on doing it "soon"
20100704 23:30:01< mordante> silene, still I added it in order to give more helpful messages to UMC developers and they might run into this issue
20100704 23:30:06< mordante> hi boucman 
20100704 23:30:10< boucman> Ivanovic: we have a good idea what the cause is and it's not trivial to fix properly...
20100704 23:30:24< Ivanovic> boucman: this sucks!
20100704 23:30:26< Ivanovic> ;)
20100704 23:30:40< boucman> Ivanovic: yes, and telling me multiple times won't help
20100704 23:30:47< Ivanovic> :)
20100704 23:31:35< boucman> i'd rather finish the bridge macros first, the artist have been waiting for them for ages, after that i'll try to have a look at those empty files in cache
20100704 23:32:25< mordante> I'm off night
20100704 23:33:06-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100704 23:34:24< silene> boucman: note that the major performance issue is just the sheer amount of macro substitutions
20100704 23:34:40< boucman> silene: really ? that's not what was diagnosed last time...
20100704 23:34:47< boucman> do unused macros have a cost ?
20100704 23:35:27< silene> boucman: not macros, macro substitutions; and it was diagnosed 10 days ago by me, and it was separately diagnosed by alink two days ago
20100704 23:36:19< boucman> I thought the problem was non-existant files making the cache index explode, that's what alink diagnosed IIRC
20100704 23:37:17-!- Deusite [~thomas@cpc1-newt26-0-0-cust245.newt.cable.virginmedia.com] has joined #wesnoth-dev
20100704 23:37:51< silene> boucman: 20100703 00:21:09< alink> I am not sure if it breaks something important but this little change http://wesnoth.pastebin.com/YSY8Bd8U seems to halve starting time when cache need rebuilding
20100704 23:39:36< boucman> silene: I'm not sure what this proves... I havn't done any tracing myself, and wasn't around at that time to get the context
20100704 23:39:47< boucman> but if removing a trace accelerates the parsing that much, go for it...
20100704 23:40:42< silene> boucman: you missed the point; the change is wrong (it disables translations and error reporting); the important thing is that the commented line is the one responsible for macro substitutions
20100704 23:41:41< boucman> that's still nor releveant... I'm sure removing all macro parsing would reduce loading time... if you use the cache and never rebuild, it would go even faster you know...
20100704 23:41:56< boucman> give me some tracing, an output from  gprof and then we'll talk
20100704 23:44:54< billynux> shadowmaster, please check the Makefile.am modifications in --v
20100704 23:45:23< CIA-87> billynux * r43917 /trunk/src/ (6 files): Separated the ana implementation into 2 new files and modified build system files.
20100704 23:45:26< billynux> ... and someone may check the SCons modifications, they are pretty trivial and should work
20100704 23:45:52< silene> boucman: http://wesnoth.pastebin.com/rCHzn8qj
20100704 23:46:03< billynux> brb, 30'
20100704 23:46:57< boucman> so we spend 63% of our time in skip_comment... and everything else is neglectable if I read correctly
20100704 23:47:58< silene> boucman: that's right, and the comments are the one created by the lines alink commented in the logs, and they are the ones that ensure that macros support translations and error reporting
20100704 23:48:33< boucman> k
20100704 23:49:39< boucman> translation is handled at macro expansion time ? that's a bit suprising...
20100704 23:50:25< Crab_> boucman: how otherwise you'd handle the case where macro is from different textdomain(s) than the place which includes it ?
20100704 23:50:39< boucman> good point
20100704 23:51:01-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 264 seconds]
20100704 23:52:11< boucman> well I don't see how to easily reduce the number of macros while keeping the current feature set... I could maybe do something to avoid the explosion due to random terrains, but I'll do it after doing the bridge macros which are most important
20100704 23:52:45< boucman> and I would really like to use macros without limits it's the only way to make complicate areas like terrain graphics usable
20100704 23:52:55< zookeeper> well, can't you just skip all macro-related textdomain stuff _if_ you're in a file which has no textdomain or including from a file which has no textdomain?
20100704 23:52:58-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100704 23:56:00< silene> it would help but not much: error reporting accounts for much more than textdomains
20100704 23:58:11-!- koda [~vittorio@95.236.230.233] has quit [Quit: "that's classy!" the method told the object]
20100704 23:59:30-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100704 23:59:48< boucman> couldn't we manage to build the error reporting line once on the first parsing of the macro and reuse it on other calls ?
--- Log closed Mon Jul 05 00:00:50 2010