--- Log opened Mon Aug 03 00:00:02 2015
20150803 00:01:51< aeth> What does it take to get Lua mainlined?
20150803 00:02:14< aeth> By the time I'm done with cleaning up my add-on, rewriting most of it, using it to define tags, etc., I'll probably have at least some functions that might be generally usable.
20150803 00:02:40< aeth> (I already have several thousand lines of code in Lua iirc)
20150803 00:03:57-!- Dacreach [637f6d94@gateway/web/freenode/ip.99.127.109.148] has left #wesnoth-dev []
20150803 00:07:33< c74d> aeth: file a pull request?
20150803 00:07:44< c74d> per <http://wiki.wesnoth.org/WesnothRepository> / <http://wiki.wesnoth.org/Git_for_Wesnoth_Crash_Course>
20150803 00:08:18< c74d> I'm not guaranteeing that the request would be granted, but if you think it's useful, might as well try
20150803 00:14:51-!- gfgtdf [~chatzilla@x50ab97f6.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 39.0/20150630154324]]
20150803 00:18:42-!- Mattaiyah_ [d8f6f374@gateway/web/freenode/ip.216.246.243.116] has joined #wesnoth-dev
20150803 00:19:05-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 256 seconds]
20150803 00:21:45-!- Mattaiyah [~Mattaiyah@216-246-243-116.cpe.distributel.net] has joined #wesnoth-dev
20150803 00:21:53-!- Mattaiyah_ [d8f6f374@gateway/web/freenode/ip.216.246.243.116] has quit [Client Quit]
20150803 00:29:58-!- Mattaiyah [~Mattaiyah@216-246-243-116.cpe.distributel.net] has left #wesnoth-dev []
20150803 00:36:13-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 246 seconds]
20150803 00:36:59-!- chaverma1 [~Adium@c-67-169-95-15.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150803 00:38:15-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20150803 00:43:40-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150803 00:44:13-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Client Quit]
20150803 00:46:13-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 252 seconds]
20150803 00:49:42-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150803 00:50:47-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20150803 00:53:51-!- Appleman1234_ [~Appleman1@KD106155094140.au-net.ne.jp] has joined #wesnoth-dev
20150803 00:55:57-!- ancestral [~ancestral@75-146-153-169-minnesota.hfc.comcastbusiness.net] has joined #wesnoth-dev
20150803 00:56:28-!- Appleman1234 [~Appleman1@KD059138023218.au-net.ne.jp] has quit [Ping timeout: 265 seconds]
20150803 00:57:00-!- ancestral_ [~ancestral@254.sub-70-197-224.myvzw.com] has joined #wesnoth-dev
20150803 01:00:29-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150803 01:00:38-!- ancestral [~ancestral@75-146-153-169-minnesota.hfc.comcastbusiness.net] has quit [Ping timeout: 260 seconds]
20150803 01:00:38-!- ancestral_ is now known as ancestral
20150803 01:02:32-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has quit [Quit: No Ping reply in 180 seconds.]
20150803 01:02:58-!- ancestral [~ancestral@254.sub-70-197-224.myvzw.com] has quit [Quit: Smell ya later!]
20150803 01:04:52-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150803 01:07:51-!- ancestral [~ancestral@254.sub-70-197-224.myvzw.com] has joined #wesnoth-dev
20150803 01:10:27-!- ancestral [~ancestral@254.sub-70-197-224.myvzw.com] has quit [Client Quit]
20150803 01:20:23-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150803 01:23:15-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 01:36:23-!- chaverma1 [~Adium@c-67-169-95-15.hsd1.ca.comcast.net] has quit [Quit: Leaving.]
20150803 01:41:59-!- [Relic] [~Relic]@2602:306:33a3:6d30:657a:e054:8393:a9de] has quit [Quit: I press the magic X and all the weirdos go away!]
20150803 01:53:51-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150803 01:55:05-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 01:59:43-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 02:00:02-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 02:01:04-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 02:19:24-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has quit [Ping timeout: 250 seconds]
20150803 02:20:47-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has joined #wesnoth-dev
20150803 02:24:21< vultraz> I wonder if I'm the only dev on Windows 10 
20150803 02:24:43< autodidact> wowee gee
20150803 02:32:48< UnwiseOwl> Any good, Vultraz?
20150803 02:33:10-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 02:37:18-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 02:37:26-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 02:46:00-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150803 02:47:20-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 02:47:48-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 02:53:38-!- [Relic] [~Relic]@2602:306:33a3:6d30:ec77:36f9:f6a5:7ae5] has joined #wesnoth-dev
20150803 02:57:17-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20150803 02:57:59-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 03:22:43< vultraz> UnwiseOwl: it is
20150803 03:38:59< shadowm> Aginor: I noticed. I'm unsure who would be the best choice to handle this PR between me and lipkab, though. It seems like lipkab is interested in overseering this tasks but I haven't asked him yet.
20150803 03:39:04< shadowm> (Hi lipkab.)
20150803 03:41:41-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 03:42:21-!- Linuturk [~Linuturk@unaffiliated/linuturk] has quit [Quit: WeeChat 1.0.1]
20150803 03:44:19-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150803 03:47:08-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 03:51:36< Aginor> shadowm: I'm hoping there will be quite a few more of them, and what I'm more interested in discussing would be how to move to SDL2 as opposed to SDL1 (I think this can be done without a lot of the graphics changes)
20150803 03:52:27< shadowm> I've discussed the matter with lipkab before and he insisted we need to use less textures.
20150803 03:52:57< shadowm> You've probably read the NSEC task's description.
20150803 03:58:02-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 03:58:28-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 03:59:49< autodidact> ?
20150803 03:59:52-!- Appleman1234__ [~Appleman1@KD036012000180.au-net.ne.jp] has joined #wesnoth-dev
20150803 04:00:58< UnwiseOwl> Not So Easy Coding
20150803 04:01:16< Aginor> I have :)
20150803 04:01:53< Aginor> I think there are different levels of what can be done there, and I don't think it need to come as one huge hit
20150803 04:02:11-!- Appleman1234_ [~Appleman1@KD106155094140.au-net.ne.jp] has quit [Ping timeout: 260 seconds]
20150803 04:27:58-!- afre [63f401a5@gateway/web/freenode/ip.99.244.1.165] has joined #wesnoth-dev
20150803 04:38:56-!- Flexfit [~Flexfit@cpe-172-74-101-237.nc.res.rr.com] has quit [Read error: Connection reset by peer]
20150803 04:39:17-!- Flexfit [~Flexfit@cpe-172-74-101-237.nc.res.rr.com] has joined #wesnoth-dev
20150803 04:39:54< afre> So I reregistered in the forums and GNA days ago to show my support. I jotted some remarks on 1.13.1. Could be useful for bug / feature / discussion. Don't feel like reformatting it atm. Should I place that info in "Users’ Forum"?
20150803 04:40:39< afre> I think that's probably the right place.
20150803 04:40:48< shadowm> Depends. Is it about a specific campaign or about the game in general?
20150803 04:41:00< shadowm> Is it exclusively about bugs that need to be fixed?
20150803 04:41:12< shadowm> Is it exclusively about possible additions to the game?
20150803 04:43:19< afre> Um. Just general comments atm. So it's in a jumble.
20150803 04:44:39< shadowm> Then Users' would be the best place I guess.
20150803 04:45:05< afre> Yeah, it isn't exclusively anything but could lead to many things haha
20150803 04:48:24< afre> Btw, which AI should I focus on testing in multiplayer. I always assumed that Experimental AI was the most current.
20150803 04:48:36< shadowm> mattsc: ^
20150803 04:49:00< shadowm> I don't know myself. I don't think the non-default options are even maintained.
20150803 04:49:41-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150803 04:49:41< afre> Oh, when there's a selection, I tend to try everything without knowing what's what.
20150803 04:49:56-!- Xara [Yangyf@2001:cc0:2020:4010:157:10b4:e611:21a7] has joined #wesnoth-dev
20150803 04:52:19< UnwiseOwl> Pretty sure the answer to the AI question is "Fred" but mattsc would know for sure
20150803 04:53:44< afre> When I post (not now; maybe later) on the forum, I'll include this question.
20150803 04:54:48-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 05:04:29-!- afre [63f401a5@gateway/web/freenode/ip.99.244.1.165] has quit [Quit: Page closed]
20150803 05:05:32-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 240 seconds]
20150803 05:15:03-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150803 05:21:50-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150803 05:26:38-!- matthew [~matthew@2602:30a:2e10:5c10:b9ac:bb0a:1eeb:f7dc] has joined #wesnoth-dev
20150803 05:26:38-!- Guest21432 [~matthew@162-225-5-193.lightspeed.dybhfl.sbcglobal.net] has quit [Read error: Connection reset by peer]
20150803 05:27:02-!- matthew is now known as Guest13613
20150803 05:52:01-!- genbattle [~genbattle@182.16.153.90] has quit [Ping timeout: 252 seconds]
20150803 05:58:50-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 05:59:01-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 06:13:44-!- [Relic] [~Relic]@2602:306:33a3:6d30:ec77:36f9:f6a5:7ae5] has quit [Quit: I press the magic X and all the weirdos go away!]
20150803 06:15:28-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 06:16:32-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 06:17:41-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 06:20:02-!- chaverma [~Chris@c-67-169-95-15.hsd1.ca.comcast.net] has quit [Quit: Leaving.]
20150803 06:20:58-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150803 06:28:33-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 06:29:22-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150803 06:36:54-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 06:42:00-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150803 06:45:05-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150803 06:52:00-!- chaverma [~Adium@c-67-169-95-15.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150803 06:55:51-!- prkc [~prkc@catv-89-134-159-162.catv.broadband.hu] has quit [Remote host closed the connection]
20150803 07:00:27-!- sailorsw_ [~sailorswi@185.94.31.252] has joined #wesnoth-dev
20150803 07:00:51-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has quit [Ping timeout: 256 seconds]
20150803 07:04:00-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 07:11:32-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 265 seconds]
20150803 07:16:24< Aginor> shadowm: I've put some of my thoughts here: http://wiki.wesnoth.org/SDL_2_%2B_OpenGL_port
20150803 07:16:47< Aginor> lipkab isn't around so I can't ping him/her
20150803 07:18:28< shadowm> He'll see your message anyway.
20150803 07:18:38< Aginor> excellent
20150803 07:22:38-!- genbattle [~genbattle@122-59-101-232.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150803 07:34:16-!- horrowin1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150803 07:39:48-!- horrowin1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has quit [Quit: horrowin1]
20150803 07:39:56-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 07:42:54-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 07:46:56-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20150803 07:48:01< Aginor> and more in that note about SDL2
20150803 07:48:02< Aginor> http://forums.wesnoth.org/viewtopic.php?f=10&t=42973
20150803 07:50:26-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20150803 07:52:38< shadowm> Moved into DD.
20150803 07:52:50-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 07:52:55< Aginor> cheers
20150803 07:52:59< Aginor> I couldn't post there
20150803 07:53:26< shadowm> I know, it's intentional.
20150803 07:53:38< Aginor> fair enough
20150803 07:54:20< Aginor> now I just need to catch up with lipkab as well
20150803 07:54:32< Aginor> does anyone know lipkab's timezone?
20150803 07:54:50< shadowm> Central Europe Time, IIRC.
20150803 07:55:32< Aginor> thanks
20150803 07:55:45< Aginor> it's not the best one, I'm GMT+13 ;)
20150803 07:55:52< shadowm> Er.
20150803 07:56:00-!- patrickB [4ec29aaa@gateway/web/freenode/ip.78.194.154.170] has quit [Ping timeout: 246 seconds]
20150803 07:56:29< shadowm> Where exactly is that this time of the year?
20150803 07:56:52< Aginor> NZ and a bunch of islands
20150803 07:57:00< Aginor> (New Zealand)
20150803 07:57:05-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 07:57:15< shadowm> But Wikipedia says it's only with DST.
20150803 07:57:37< shadowm> And it's currently winter.
20150803 07:57:37< Aginor> maybe I am confused
20150803 07:57:57< shadowm> Let me rephrase the question: exactly where in the Pacific are you? :p
20150803 07:58:24< Aginor> New zealand
20150803 07:58:38< Aginor> in my defence, I'm not a native so I'm allowed to be confused :D
20150803 07:58:41< shadowm> Which toolchain did you use to build with SDL 2?
20150803 07:59:05< Aginor> scons
20150803 07:59:08< Aginor> cmake is lacking
20150803 07:59:17< Aginor> I'll give it some TLC in a while
20150803 07:59:52< Aginor> I'll probably move over to using cmake as theprimary one and use eclipse cdt
20150803 08:01:55< shadowm> scons: *** [build/release/about.o] AttributeError : 'NoneType' object has no attribute 'get_contents'
20150803 08:02:04< shadowm> Okay, this is a scons issue I've never seen before.
20150803 08:02:33< shadowm> shadowm@nanacore:~/src/wesnoth/sdl2-build% scons -Y .. -j8 build=release sdl2=yes nls=no wesnoth           
20150803 08:02:38< Aginor> ummm
20150803 08:02:38< shadowm> loonycyborg: Any idea what this error means?
20150803 08:03:03< Aginor> I did "scons -j8 build=debug sdl2=true"
20150803 08:03:28< loonycyborg> Aginor: try without -Y
20150803 08:03:32< shadowm> yes/true mean the same thing, and I get the same result with build=debug.
20150803 08:03:46< shadowm> loonycyborg: I don't want to, also it's me you should be highlighting. :\
20150803 08:03:55< loonycyborg> ohwhoops
20150803 08:04:27< shadowm> Not too long ago I used -Y for a crosscompiler build and I didn't hit anything like this.
20150803 08:04:43< loonycyborg> hmm I remember that issue, it went away when I removed sconsign.dblite from source dir
20150803 08:05:00< loonycyborg> it seems to be caused by scons upgrade
20150803 08:05:20< loonycyborg> and it still using sconsign.dblite from source dir somehow
20150803 08:05:50< shadowm> I haven't upgraded in over a month. :\
20150803 08:06:07< shadowm> You were wondering the other day why I preferred to use cmake instead of scons -Y.
20150803 08:06:18-!- chaverma [~Adium@c-67-169-95-15.hsd1.ca.comcast.net] has quit [Quit: Leaving.]
20150803 08:07:33< shadowm> loonycyborg: There is no such file.
20150803 08:07:45< loonycyborg> in build/
20150803 08:07:56< shadowm> You said source dir, ugh.
20150803 08:08:20< loonycyborg> I mean source with respect to current dir
20150803 08:08:39< loonycyborg> hmm or no
20150803 08:08:46-!- hay207 [~hay207@41.34.12.122] has joined #wesnoth-dev
20150803 08:08:53< loonycyborg> I get it now too from /tmp/wesnoth
20150803 08:08:58< shadowm> Now it works.
20150803 08:09:16< shadowm> After removing both ../build/yaddayadda and ./build/yaddayadda.
20150803 08:11:59< loonycyborg> very interesting. so it's not version upgrade that is responsiblefor this database brainfart
20150803 08:15:24< shadowm> Aginor: I get some compiler warnings.
20150803 08:15:58< shadowm> Which is interesting seeing as how the CI build passed.
20150803 08:17:33< shadowm> http://pastebin.com/266CGVmK
20150803 08:17:51< shadowm> Yeah, scrolling is really trippy.
20150803 08:19:41-!- genbattle [~genbattle@122-59-101-232.jetstream.xtra.co.nz] has quit [Ping timeout: 265 seconds]
20150803 08:19:48< Aginor> I got the impression that those warnings already existed, based on how the code's written
20150803 08:19:50< shadowm> Htokeys also seem completely dysfunctional -- e.g. F1 brings up the command box as if I pressed : instead, and Ctrl+Q opens the help as if I had pressed F1 instead.
20150803 08:19:58< Aginor> and marked them down as "fix in the future"
20150803 08:20:12< Aginor> (ssize_t is compared with int)
20150803 08:20:18< shadowm> Oh right, SDL 2-specific code isn't built by travis yet.
20150803 08:20:33< Aginor> (size_t shouldn't be used there in the first place!)
20150803 08:20:44< shadowm> Of course one can't get warnings from #ifdef'd-out code.
20150803 08:20:50< Aginor> indeed
20150803 08:21:54< Aginor> I don't think I've introduced new warnings, but I haven't fixed existing warnings yet
20150803 08:22:49< Aginor> my plan was to do that in future pull requests where I refine things and aim towards experience parity if possible
20150803 08:23:27 * shadowm nods.
20150803 08:25:13-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150803 08:25:13< Aginor> well, insofar I have a plan
20150803 08:27:18< loonycyborg> Aginor: what did you change with respect to graphics rendering?
20150803 08:27:51< shadowm> lipkab: Can we agree to never use this obnoxious ttype naming convention outside GUI2 btw?
20150803 08:28:10< Aginor> loonycyborg: nothing?
20150803 08:28:36< loonycyborg> was just checking, I've read sdl2 porting guide
20150803 08:28:56< Aginor> https://github.com/aginor/wesnoth/commit/aae2ad77e6d2fc1d746248072688fbcb6adc6a57
20150803 08:28:58< loonycyborg> iirc it was recommending to switch to opengl or something like that
20150803 08:28:58< shadowm> tI tdon't twant tto tname tall tclass tor tstruct ttypes tlike tthis. CI'd CRather CUse CThis CIf CI CAbsolutely CHad CTo, CThough.
20150803 08:29:19< Aginor> loonycyborg: yes, but that's a future problem :D
20150803 08:29:27< Aginor> step 1: make it compile
20150803 08:29:36< Aginor> step2: fix issues
20150803 08:29:47< shadowm> lipkab: (And yes, I'm aware of CVideo. I ignore it because it's been there since version 0.1's immediate predecessor.)
20150803 08:29:49< Aginor> step 3: Add clipboard (not implemented)
20150803 08:29:50< loonycyborg> ah yes cool
20150803 08:30:07< Aginor> step 4: Investigate rendering changes
20150803 08:30:17< Aginor> Step 5: profit!!!
20150803 08:30:28< loonycyborg> afaik ttype convention comes from pascal or something
20150803 08:30:39< loonycyborg> and it being in wesnoth is entirely mordante's fault
20150803 08:30:50-!- Guest13613 [~matthew@2602:30a:2e10:5c10:b9ac:bb0a:1eeb:f7dc] has quit [Remote host closed the connection]
20150803 08:30:58< Aginor> is there a common coding style?
20150803 08:31:12< Aginor> I've seen mixed usage of braces in different parts of the code...
20150803 08:31:20-!- genbattle [~genbattle@122-59-101-232.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150803 08:31:38< shadowm> loonycyborg: Yes, but it used to be restricted to ttext and the GUI2 type hierarchy and very few other types. Then the SDL_gpu project happened.
20150803 08:31:47-!- matthew [~matthew@162-225-5-193.lightspeed.dybhfl.sbcglobal.net] has joined #wesnoth-dev
20150803 08:32:01< shadowm> I say keep it confined to GUI2 land where everything is weird.
20150803 08:32:06< loonycyborg> who was mentor of that project?
20150803 08:32:11< shadowm> mordante, obviously.
20150803 08:32:11-!- matthew is now known as Guest85591
20150803 08:32:18< loonycyborg> not surprised totally
20150803 08:34:06-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150803 08:34:39< shadowm> Aginor: No, there isn't a consistent coding style, although GUI2 (src/gui/ and src/.clang-format, which is intended to be used with the clang-format tool) serves as an adequate reference most of the time -- except for the C++11/14 conditional symbols, the use of FOREACH over BOOST_FOREACH (which is just a particular case of the language conditionals, really), tthe tfunny ttype tnames, and the excess use of templates.
20150803 08:35:29< Aginor> shadowm: fair enough, thank you
20150803 08:35:37< loonycyborg> BOOST_FOREACH is shouldalways be used directly
20150803 08:36:06< loonycyborg> we used #define foreach BOOST_FOREACH and it backfired horribly
20150803 08:36:23< Aginor> any thoughts about typedefs?
20150803 08:36:26< shadowm> The C++11/14 conditionals used in GUI2 are symbols like OVERRIDE (is the override keyword with C++11, nothing otherwise), FINAL (idem with final), FOREACH (is BOOST_FOREACH with C++03, range for otherwise), AUTO (auto with C++11/14, no idea what otherwise), and that's it I believe.
20150803 08:36:58-!- Appleman1234_ [~Appleman1@KD111107184074.au-net.ne.jp] has joined #wesnoth-dev
20150803 08:37:12< shadowm> Oh and GUI2 tries to emulate `enum class` in C++03 in some places, and that makes some code look ugly.
20150803 08:38:08< Aginor> I'm sure I'll have more opportunity to look at this more closely
20150803 08:38:11< shadowm> Aginor: We generally avoid frivolous typedefs like `typedef std::pair<int, int> pair_of_ints`.
20150803 08:39:03< Aginor> shadowm: fair enough, but acceptable with proper use?
20150803 08:39:13< loonycyborg> why it wouldn't be acceptable?
20150803 08:39:33< Aginor> I'm more thinking of the linux kernel where raw types should always be used :D
20150803 08:39:39< shadowm> I think a good rule of thumb regarding where to use a global, class or namespace-level typedef is whether it'd be possible and desirable to change the underlying type without changing the code's semantics.
20150803 08:39:57< shadowm> That's what I do in my own code at least.
20150803 08:40:03< loonycyborg> wesnoth isn't linux kernel
20150803 08:40:05< Aginor> loonycyborg: because it's hiding the original types
20150803 08:40:18-!- Appleman1234__ [~Appleman1@KD036012000180.au-net.ne.jp] has quit [Ping timeout: 250 seconds]
20150803 08:40:19< shadowm> For example, typedef std::map<std::string, std::string> string_map (it exists).
20150803 08:40:53< Aginor> fair enough, that gives me something to go by
20150803 08:40:56< shadowm> Okay, bad example, but that's good example of the other rule of thumb: how often am I going to have to type out the full name of this template type?
20150803 08:41:19< Aginor> loonycyborg: I realise that, I'm asking because I'm trying to figure out how things are done so my code will fit in :)
20150803 08:41:36< shadowm> A better example would be something like typedef std::vector<some non-POD type here> foo_list.
20150803 08:41:42< loonycyborg> I think it's a verygood idea to typedef stuff like boost.bimapand boost.multi-index
20150803 08:43:03< shadowm> In this case perhaps at some later point I decide to replace the item type with another type that has the exact same API but is better suited to the task.
20150803 08:43:35< shadowm> In that case I only need to replace the item type in that typedef instead of all over the place.
20150803 08:43:56< Aginor> yeah
20150803 08:44:00< Aginor> makes sense
20150803 08:44:05< loonycyborg> boost::bimaps::bimap<boost::bimaps::unordered_multiset_of<socket_ptr>, boost::bimaps::unordered_multiset_of<std::string>,boost::bimaps::set_of_relation<>,boost::bimaps::with_info<room_ptr>
20150803 08:44:05< Aginor> try to decouple
20150803 08:44:18< loonycyborg> totally need a typedef for thing like that
20150803 08:45:46< Aginor> yeah
20150803 08:47:24-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has joined #wesnoth-dev
20150803 08:47:36-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 08:47:43< shadowm> Anyway, I'm off. If you have any other questions (for most of which the answer will be inevitably "we don't have hard rules for this"), I'll try to answer them later.
20150803 08:48:32-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20150803 08:48:46< Aginor> fair enough
20150803 08:48:51< Aginor> thank you for your time
20150803 08:51:35-!- cpf [~cpf@ip-188-118-61-46.reverse.destiny.be] has joined #wesnoth-dev
20150803 08:51:41-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 08:53:03-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 256 seconds]
20150803 08:55:41-!- genbattle_ [~genbattle@115-188-157-223.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150803 08:56:26-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20150803 08:57:06-!- genbattle [~genbattle@122-59-101-232.jetstream.xtra.co.nz] has quit [Ping timeout: 240 seconds]
20150803 08:59:26-!- boucman_work [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20150803 08:59:26-!- boucman_work [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Changing host]
20150803 08:59:26-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150803 09:37:24-!- irker807 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150803 09:37:24< irker807> wesnoth: Charles Dang wesnoth:master d10b206a8bd2 / src/game_launcher.cpp: Some on-launch stderr formatting http://git.io/vOl9x
20150803 09:41:35-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 09:42:06-!- Guest85591 [~matthew@162-225-5-193.lightspeed.dybhfl.sbcglobal.net] has quit [Remote host closed the connection]
20150803 09:45:51-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 09:45:59-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 09:53:10-!- Appleman1234_ is now known as Appleman1234
20150803 09:59:51-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
20150803 10:03:41-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20150803 10:04:25-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20150803 10:16:27-!- travis-ci [~travis-ci@ec2-54-159-34-72.compute-1.amazonaws.com] has joined #wesnoth-dev
20150803 10:16:28< travis-ci> wesnoth/wesnoth#6936 (master - d10b206 : Charles Dang): The build has errored.
20150803 10:16:28< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/73862184
20150803 10:16:28-!- travis-ci [~travis-ci@ec2-54-159-34-72.compute-1.amazonaws.com] has left #wesnoth-dev []
20150803 10:17:59< vultraz> dammit traaaavis
20150803 10:18:24-!- sailorsw_ [~sailorswi@185.94.31.252] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150803 10:21:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150803 10:22:55< Aginor> is there a reason all of the widgets are using size_t a lot as an unsigned integer? would anyone object to me changing the types to "int" or "unsigned int" and made a pull request or is it a waste of time?
20150803 10:24:12< vultraz> I'm not sure, but if it's an improvement it should get accepted
20150803 10:25:06< Aginor> I'm not sure it counts as an improvement, it's more of a "tidying up". size_t is normally used to indicate byte sizes for things...
20150803 10:25:55< Aginor> here it's used for ids, coordinates (some, but not all), and I don't know what exactly
20150803 10:26:00< Aginor> might look at it tomorrow
20150803 10:32:52-!- matthew [~matthew@162-225-5-193.lightspeed.dybhfl.sbcglobal.net] has joined #wesnoth-dev
20150803 10:33:12-!- matthew is now known as Guest80367
20150803 10:41:34-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20150803 10:47:48-!- genbattle_ [~genbattle@115-188-157-223.jetstream.xtra.co.nz] has quit [Ping timeout: 244 seconds]
20150803 10:53:19< loonycyborg> Aginor: it might make sense for them to be size_t if they're at some point used for array indexing
20150803 10:54:50-!- Guest80367 [~matthew@162-225-5-193.lightspeed.dybhfl.sbcglobal.net] has quit [Remote host closed the connection]
20150803 10:58:34-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150803 11:17:46-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150803 11:27:15-!- Xara [Yangyf@2001:cc0:2020:4010:157:10b4:e611:21a7] has quit [Read error: Connection reset by peer]
20150803 11:35:57-!- Xara [Yangyf@2001:cc0:2020:4010:157:10b4:e611:21a7] has joined #wesnoth-dev
20150803 11:44:45-!- Appleman1234 [~Appleman1@KD111107184074.au-net.ne.jp] has quit [Ping timeout: 252 seconds]
20150803 11:47:26-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 11:51:50-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 12:22:00-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150803 12:28:32-!- Appleman1234 [~Appleman1@KD036012009066.au-net.ne.jp] has joined #wesnoth-dev
20150803 12:46:50-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 12:50:33-!- horrowin1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150803 12:51:12-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 12:51:28-!- Appleman1234 [~Appleman1@KD036012009066.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20150803 13:04:26-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150803 13:04:36-!- quentinp [~quentin@91.121.196.50] has joined #wesnoth-dev
20150803 13:05:26-!- Xara2 [Yangyf@2001:cc0:2020:4010:74e1:5dbd:cfc0:ceae] has joined #wesnoth-dev
20150803 13:06:18-!- cpf [~cpf@ip-188-118-61-46.reverse.destiny.be] has quit [Ping timeout: 260 seconds]
20150803 13:08:51-!- Xara [Yangyf@2001:cc0:2020:4010:157:10b4:e611:21a7] has quit [Ping timeout: 244 seconds]
20150803 13:09:11-!- irker807 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150803 13:11:12-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has quit [Ping timeout: 264 seconds]
20150803 13:15:13-!- cpf [~cpf@ip-188-118-61-46.reverse.destiny.be] has joined #wesnoth-dev
20150803 13:15:33-!- gandaro [~gandaro@wikipedia/Gorlingor] has joined #wesnoth-dev
20150803 13:15:38-!- Appleman1234 [~Appleman1@KD036012009066.au-net.ne.jp] has joined #wesnoth-dev
20150803 13:16:02-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150803 13:16:30< quentinp> since the call for help, I decided to try if I could fix one bug - the first step was to build and run wesnoth on my OS X machine
20150803 13:16:54< quentinp> I was quite surprised to see that building wesnoth was quick enough (about 10 minutes) with all dependencies installed
20150803 13:17:02< ancestral> quentinp: Homebrew?
20150803 13:17:04< quentinp> but running it was more difficult
20150803 13:17:33< quentinp> ancestral: I use homebrew, but I built it from the git sources
20150803 13:17:56< ancestral> Did you use Homebrew to build Wesnoth?
20150803 13:17:56< quentinp> when running, ./wesnoth, I found out that it looked for fonts.cfg in /usr/local/share/wesnoth/data/hardwired/fonts.cfg
20150803 13:18:03< quentinp> ancestral: no, I used scons
20150803 13:18:13< ancestral> Which is what Homebrew does
20150803 13:18:29< quentinp> ok - I used scons directly
20150803 13:18:34 * ancestral nods
20150803 13:18:40< quentinp> http://wiki.wesnoth.org/CompilingWesnothOnMacOSX
20150803 13:19:01< quentinp> when passing --data-dir data to wesnoth, it looked for /Users/quentin/Software/wesnoth/data/data/hardwired/fonts.cfg
20150803 13:19:12< quentinp> I finally managed to get wesnoth running by passing --data-dir .
20150803 13:19:20< quentinp> is this intended?
20150803 13:23:50-!- gandaro [~gandaro@wikipedia/Gorlingor] has quit [Quit: Leaving]
20150803 13:27:04< quentinp> should I just wait a few hours, or should I head to the forums for that question?
20150803 13:28:27< loonycyborg> quentinp: I don't think it's intended
20150803 13:29:07< loonycyborg> running wesnoth from git checkout root shouldalways use all data files from that checkout
20150803 13:29:08< ancestral> quentinp: The two ways I’ve built Wesnoth for OS X
20150803 13:29:57< ancestral> Through Homebrew: http://r.wesnoth.org/p577235
20150803 13:30:22< ancestral> Or git pull the project, download the Mac Compile Stuff, and build in Xcode
20150803 13:31:02-!- scorpion [~scorpion@46.166.190.130] has quit [Ping timeout: 260 seconds]
20150803 13:31:26< ancestral> But, just like Perl, there’s more than one way to do it
20150803 13:32:11< ancestral> quentinp: This is the Mac Compile Stuff: https://sourceforge.net/projects/wesnoth/files/unofficial/Mac%20Compile%20Stuff/
20150803 13:32:40< ancestral> (Note that 1.11 is for 1.11 and 1.12)
20150803 13:32:42< quentinp> thanks for the links
20150803 13:33:07< loonycyborg> running ./wesnoth from git checkout root is the way that works on linux
20150803 13:33:18-!- Xara [~Yangyf@210.77.22.55] has joined #wesnoth-dev
20150803 13:33:27< loonycyborg> if it doesn't on mac due to some platform specific code perhaps then it's a bug
20150803 13:33:28< ancestral> Honestly, it’s a breeze going through Homebrew, and if you use the ccache flag, you can pull from the project without having to redownload everything
20150803 13:33:31< loonycyborg> or missiongfeature
20150803 13:33:38< ancestral> and rebuild
20150803 13:34:12< ancestral> quentinp: Just curious, what version of OS X are you on?
20150803 13:34:14< quentinp> I'm a long-time Linux user, which is why I looked for a way to build from the root folder and edit in vim, not XCode
20150803 13:34:30< quentinp> ancestral: 10.10.4, it seems
20150803 13:34:35< ancestral> Sure
20150803 13:34:58< quentinp> sounds like the homebrew way could work well
20150803 13:35:27< quentinp> the last time I changed a file there, I broke all updates, which is why I didn't consider this option yet
20150803 13:35:48< quentinp> I guess I'll check why it looks for /usr/local first
20150803 13:35:56< quentinp> then use a different way to build wesnoth
20150803 13:36:14-!- Xara2 [Yangyf@2001:cc0:2020:4010:74e1:5dbd:cfc0:ceae] has quit [Ping timeout: 244 seconds]
20150803 13:36:31< ancestral> mattsc and I are good resources if you have questions. StandYourGround has had success too. And for general homebrew questions, iceiceice knows a lot about it
20150803 13:36:38< quentinp> thanks!
20150803 13:36:40< ancestral> :)
20150803 13:37:21< ancestral> Always great having another OS X builder (even if you’re going through the command line)
20150803 13:42:36< loonycyborg> quentinp: actually for "./wesnoth" working without extra options we need to detect directory where the binary is located
20150803 13:42:45< loonycyborg> perhaps it just isn't implemented on macos
20150803 13:42:59< quentinp> loonycyborg: do you know where that code lives?
20150803 13:43:14< loonycyborg> before that functionality existed the way to run it was "./wesnoth ."
20150803 13:43:24< quentinp> wesnoth.cpp:947, the call to get_exe_dir ?
20150803 13:44:08< loonycyborg> yes
20150803 13:44:37< quentinp> indeed loonycyborg that's the issue
20150803 13:44:40< quentinp> it's empty on OS X
20150803 13:47:36-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150803 13:48:12-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150803 13:48:37-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Client Quit]
20150803 13:49:34< quentinp> loonycyborg: what are the possible values for detecting OSes? I see at least _WIN32 and __APPLE__
20150803 13:49:54< quentinp> the get_exe_dir code uses '#ifndef _WIN32' to access /proc which does not exist on OS X
20150803 13:50:11< loonycyborg> I guess #ifdef __APPLE__
20150803 13:50:13< loonycyborg> not sure though
20150803 13:50:42< quentinp> loonycyborg: I mean, what is the value for linux?
20150803 13:50:49< quentinp> not windows != linux
20150803 13:51:23< loonycyborg> code assumes that any non windows is a posix os
20150803 13:51:44< loonycyborg> and assumes that it'llwork on linux and fail on non-linux
20150803 13:51:49-!- gfgtdf [~chatzilla@x50ab97f6.dyn.telefonica.de] has joined #wesnoth-dev
20150803 13:51:56< loonycyborg> so it doesn't do explicit checking for linux
20150803 13:51:57< quentinp> yeah but /proc/self/exe is not posix :)
20150803 13:52:36< loonycyborg> yes so basically it's check for linux is the very act of looking for /proc/self/exe
20150803 13:52:55< gfgtdf> i wonder whether it is works to have a [random] in conditional wml like in [if][random] chance = 70% [/random] [then] could mostlikelkey easy be implementad with iceiceices new wml_conditional lua table.
20150803 13:53:03< gfgtdf> s/works/worth
20150803 13:53:20< loonycyborg> *its
20150803 13:53:39< gfgtdf> or also [event] [filter_condition] [random]
20150803 13:57:27< quentinp> loonycyborg: https://github.com/wesnoth/wesnoth/pull/440
20150803 13:58:00-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150803 13:58:31< zookeeper> gfgtdf, for [event] [filter_condition] it could get a bit weird. would it randomize every time the event is evaluated for possible triggering? what about if you have several events of the same type, or if one event fires another, etc.
20150803 13:58:53< zookeeper> i'm sure it could be specified, but it still seems like a potential can of worms
20150803 13:59:10< loonycyborg> quentinp: ty
20150803 14:00:31< gfgtdf> zookeeper: yes it would be randomize every time the event is evaluated for possible triggering, unrelated to woehther teh event was ttriggered by the engine or by other wml code
20150803 14:01:56< gfgtdf> zookeeper: so [event] name = "moveto", first_time_only = no [filter_conditioan] .. percentage = 50 .. [/filter_conditioan][/event] would be an event that is fired at every move with a 50% chance.
20150803 14:02:29-!- ancestral [~ancestral@84.sub-70-197-199.myvzw.com] has joined #wesnoth-dev
20150803 14:02:34< gfgtdf> zookeeper: i persoanlly think the {RANDOM} + filter variable is quiet hard to read.
20150803 14:06:08< zookeeper> i find it pretty easy, considering how it's one of the most common constructs :p
20150803 14:08:05< quentinp> ancestral: it would be nice to run travis tests on OS X, too: http://docs.travis-ci.com/user/multi-os/ - what do you think?
20150803 14:08:59< ancestral> Yes
20150803 14:09:52< zookeeper> gfgtdf, for [if] conditions, i'd think it'd make more sense to have the [random] condition support all the comparison keys that [variable] does. so, you'd use it like this:
20150803 14:09:53< zookeeper> [if] [random] rand=0..100 greater_than=50 [/random] [then] ...
20150803 14:10:15-!- Flexfit [~Flexfit@cpe-172-74-101-237.nc.res.rr.com] has quit [Read error: Connection reset by peer]
20150803 14:10:40< ancestral> quentinp: I don’t have much experience doing this, but sure
20150803 14:10:48< quentinp> ok
20150803 14:10:48< zookeeper> or [if] [random] rand=Spearman,Bowman,Horseman equals=Bowman [/random] [then] ... etc
20150803 14:11:57< gfgtdf> zookeeper: hm i actually have never seen somethign like {RANDOM } +  consitional variable with rand = non-interger values
20150803 14:12:14-!- Flexfit [~Flexfit@cpe-172-74-101-237.nc.res.rr.com] has joined #wesnoth-dev
20150803 14:12:16< gfgtdf> zookeeper: also your example with greater= is doesnt give a 50% chance
20150803 14:12:50< quentinp> ancestral: it's nothing complicated: 1/ send an email to travis to enable this beta feature 2/ enable os x but allow failures there 3/ fix commands in .travis.yml for OS X (which are only shell commands) 4/ stop allowing failures :)
20150803 14:13:20< ancestral> Cool
20150803 14:13:43< zookeeper> gfgtdf, sure, i don't recall any such uses
20150803 14:16:01< quentinp> one difficult part could be replacing xvfb
20150803 14:16:05< zookeeper> gfgtdf, i wouldn't mind your idea as such, but it's really just a shortcut for one very specific kind of randomization.
20150803 14:16:33-!- Tylertwo [~a@unaffiliated/tylertwo] has joined #wesnoth-dev
20150803 14:16:41< zookeeper> if people have tired of {RANDOM}+[if] combos then they can just use a trivial shortcut macro
20150803 14:18:11-!- Tylertwo [~a@unaffiliated/tylertwo] has quit [Client Quit]
20150803 14:19:39-!- Xara2 [~Yangyf@210.77.22.55] has joined #wesnoth-dev
20150803 14:19:43-!- ancestral [~ancestral@84.sub-70-197-199.myvzw.com] has quit [Read error: Connection reset by peer]
20150803 14:20:45-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150803 14:23:04-!- Xara [~Yangyf@210.77.22.55] has quit [Ping timeout: 272 seconds]
20150803 14:23:21< zookeeper> gfgtdf, on a related note, you know what would be great? the ability to store _one_ (or a specified number of) rando unit or location with [store_unit] and [store_locations]
20150803 14:23:27< zookeeper> random, even
20150803 14:24:08< zookeeper> there's so many times you want to, say, spawn a unit on a random location (limited by a SLF) or pick a random unit (limited by a SUF)
20150803 14:24:43< zookeeper> and you always have to store all possibilities, get a random index, and then pick the unit/location from that index
20150803 14:27:54< zookeeper> it'd be nice if you could just stick a random=4 in your [store_unit] and have the result contain 4 random units out of all matching units
20150803 14:35:43< gfgtdf> zookeeper: hm i dont think that could be imepleted in lua.
20150803 14:35:53< gfgtdf> zookeeper: at least not easily
20150803 14:35:54-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has joined #wesnoth-dev
20150803 14:36:08< gfgtdf> zookeeper: or wait
20150803 14:36:34< gfgtdf> zookeeper: i think [store_unit] is actually implemnted in lua
20150803 14:36:41< quentinp> is there a way to relaunch a travis build? https://travis-ci.org/wesnoth/wesnoth/builds/73896499
20150803 14:36:53< quentinp> two builds failed because no of no output for 10 minutes
20150803 14:37:21< quentinp> in apt-get install and apt-get update, respectively
20150803 14:39:29< aquileia> quentinp: Before you write an email to travis - I think there's another optional feature we wanted to ask from them (it seems they have a few machines running an updated ubuntu)
20150803 14:41:08< aquileia> quentinp: Devs can restart builds (which I did for your PR)
20150803 14:41:56< mattsc> gfgtdf: yes, it is: wesnoth.get_units()
20150803 14:42:11< quentinp> aquileia: thanks for the restart!
20150803 14:44:22< mattsc> gfgtdf: did you see PR #438?  It’s one of those things for which I had to turn off strict compiling in Xcode, but I have no idea if that is the correct way of fixing it.  So I’d appreciate if you could have a look at it.
20150803 14:46:35< quentinp> aquileia: Ubuntu 14.04 does not appear to be available on Travis
20150803 14:46:43-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has joined #wesnoth-dev
20150803 14:46:52< gfgtdf> mattsc: currently that object only lives on the stack so we dont need a virtual destrucor but i dont see disavantages of having one.
20150803 14:47:21< gfgtdf> mattsc: feel free to merge.
20150803 14:47:46-!- [Relic] [~Relic]@2602:306:33a3:6d30:a428:b081:a078:8eaf] has joined #wesnoth-dev
20150803 14:47:55< quentinp> aquileia: http://next.travis-ci.com/ suggest that there's a Ubuntu 14.04 beta, is this what you were thinking about?
20150803 14:48:37< aquileia> I think that's it (was looking through the logs for iceiceice's comment on it)
20150803 14:49:58< aquileia> It would save us updateing part of the build environment
20150803 14:51:20-!- irker584 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150803 14:51:20< irker584> wesnoth: Patrick Bechon wesnoth:master deafe6e0f628 / src/synced_context.hpp: Fix a compilation error with XCode Version 6.2 on OSX 10.9 http://git.io/vO4zP
20150803 14:51:20< irker584> wesnoth: Charles Dang wesnoth:master d2375c07b6a1 / src/synced_context.hpp: Merge pull request #438 from PBechon/master http://git.io/vO4zX
20150803 14:52:51< mattsc> gfgtdf: Thanks.  [Just FYI, it produces 33 warning messages in Xcode which is why strict compiling does not work.]
20150803 14:53:12-!- jjnm [~jjnm@d72-39-242-29.home1.cgocable.net] has joined #wesnoth-dev
20150803 14:53:19< mattsc> vultraz: thanks
20150803 14:53:29< vultraz> no prob
20150803 14:54:17< gfgtdf> mattsc: most likeley becasue that file is included in ~33 other files.
20150803 14:55:19< mattsc> gfgtdf: right;  I didn’t mean to ask or “complain” about it, just explaining what happened.
20150803 14:55:58< aquileia> quentinp, shadowm: I guess a repo admin has to send the email to request these two travis beta features?
20150803 14:56:16< mattsc> gfgtdf: what I don’t know is why Xcode often produces warnings when other compilers or build systems do not.  I guess they just have different levels of things that they throw warnings for?
20150803 14:57:59< vultraz> Could be the compiler
20150803 14:58:00< aquileia> shadowm: we're speaking about additional OSX builds (http://docs.travis-ci.com/user/multi-os/)  & Ubuntu 14.04 Support (http://next.travis-ci.com/)
20150803 14:58:25< gfgtdf> mattsc: im not sure, i thought that other compiler also have this wanroing about missing virtual desstructor
20150803 14:58:27< quentinp> aquileia, shadowm: I guess so.
20150803 14:58:48< vultraz> I've gotten no such warning here
20150803 14:58:54< gfgtdf> mattsc: for example previousl to this commit https://github.com/wesnoth/wesnoth/commit/8943d742a86f78302d35cd3f32b699e99790e4ce i also git a warning about missing virutla dstructor
20150803 14:59:03< vultraz> (tdm gcc 5.1.0)
20150803 14:59:27-!- Xudo [bce89833@gateway/web/freenode/ip.188.232.152.51] has joined #wesnoth-dev
20150803 15:00:06< aquileia> That's why I pinged shadowm
20150803 15:00:40< vultraz> mattsc: BTW, didn't you and ancestral manage to fix that newline thing on os x awhile back?
20150803 15:00:43< mattsc> gfgtdf: okay; well, if there’s any information I can provide to help with these things, let me know.  But I assume we’ll just consider this issue solved?
20150803 15:01:12< mattsc> vultraz: that’s the empty new lines not showing up in messages thing?
20150803 15:01:38< mattsc> s/empty new lines/empty lines
20150803 15:03:44< mattsc> quentinp: when you build directly with scons on OS X, does that give you a Wesnoth.app directory, with the executable in Contents/MacOS/ ?
20150803 15:04:46< vultraz> mattsc: yes
20150803 15:04:50< mattsc> I’m asking because for builds with Xcode, you can go into the directory and start from the CL without needin data-dir
20150803 15:05:03< mattsc> Could previous to your PR already, I mean
20150803 15:06:01< mattsc> quentinp: that means that I cannot test your PR (other than making sure that it did not break what already worked on my system)
20150803 15:06:35< quentinp> mattsc: I don't see Wesnoth.app in my wesnoth checkout, no
20150803 15:07:45< mattsc> quentinp: ok  (that’s what I expected)
20150803 15:07:54< quentinp> I don't think wesnoth got installed by scons, it only got built
20150803 15:08:16< quentinp> mattsc: not sure why you cannot test my PR
20150803 15:08:17< mattsc> vultraz: I am reasonably sure that that bug went away when I compiled a newer version of the pango libs
20150803 15:08:28< quentinp> scons jobs=3 gettextdir=`brew --prefix gettext` && ./wesnoth
20150803 15:08:28< mattsc> vultraz: so it is fixed in 1.13, but not in 1.12
20150803 15:09:31< mattsc> quentinp: let me be more precise:  I cannot test it with the way how I usually build Wesnoth (using Xcode), since it worked for that already
20150803 15:09:50< gfgtdf> does anyone know what the result of (long + unsigned int) is ? Is it 'long' 'ungigned int' or 'unsigned long' or does it depend on the sized of long and int ?
20150803 15:09:51< quentinp> mattsc: sure :)
20150803 15:11:52< mattsc> I don’t even have homebrew installed at the moment.  (I’m used to using Macports)
20150803 15:12:33< mattsc> Anyways, I should probably do that at some point anyway, but right now I need to be off.  Thanks for helping out.
20150803 15:12:33< quentinp> mattsc: if you installed gettext without homebrew, you don't need to pass gettextdir
20150803 15:12:41< quentinp> sure, have a nice day
20150803 15:13:01< mattsc> quentinp: okay, I see
20150803 15:13:04< mattsc> thanks, you too
20150803 15:13:42-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20150803 15:15:14< aquileia> quentinp: The discussion about Ubuntu 14.04 was here: http://www.wesnoth.org/irclogs/2015/07/%23wesnoth-dev.2015-07-27.log @ 19:35
20150803 15:15:29< aquileia> Just for reference
20150803 15:17:20< quentinp> thanks
20150803 15:21:27< Ravana_> is there any reason to continue disallowing [switch] [case] value=true ?
20150803 15:22:43-!- Xara2 [~Yangyf@210.77.22.55] has quit [Read error: Connection reset by peer]
20150803 15:25:59< Ravana_> actually seems that is fixed already
20150803 15:28:15-!- jjnm [~jjnm@d72-39-242-29.home1.cgocable.net] has quit [Quit: Leaving]
20150803 15:30:56< gfgtdf> zookeeper: you know where helper.shuffle is used ?
20150803 15:40:21-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 15:41:06< gfgtdf> mattsc: you have an opinion on https://github.com/wesnoth/wesnoth/pull/441/files ?
20150803 15:44:39-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 15:49:27-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150803 15:49:49< Xudo> gfgtdf: can you tell me what is wrong with default math.random in lua?
20150803 15:50:16< vultraz> Xudo: causes out of sync errors
20150803 15:50:47< vultraz> mattsc: do you happen to know which newer version of pango that was?
20150803 15:50:50< mattsc> gfgtdf: I don’t use helper.shuffle, so I won’t have any issues with the change.  I don’t know how many peolpe might, but it probably does make sense to have it synced on general grounds.
20150803 15:51:05< mattsc> gfgtdf: what’s the difference between wesnoth.random() and helper.rand()
20150803 15:51:07-!- prkc [~prkc@catv-89-134-159-162.catv.broadband.hu] has joined #wesnoth-dev
20150803 15:51:28< Xudo> I don't see anytning about syncing in proposed function
20150803 15:51:55< Xudo> or it is incapsulated in random_new::generator->next_random() ?
20150803 15:52:24< mattsc> Xudo, vultraz: on the other hand, in Lua AI move evaluations, you need to use math.random, as using the synced RNG would could OOS.  :)
20150803 15:53:21< mattsc> vultraz: 1.36.3
20150803 15:53:36< vultraz> hm...
20150803 15:53:53< vultraz> Someone tested on windows with 1.37 and he said it didn't fix it :/
20150803 15:54:18< mattsc> vultraz: and cairo 1.12.16
20150803 15:54:31< vultraz> I believe he used 1.14
20150803 15:54:53< vultraz> Leads me to think the problem might be more deep set on windows :(
20150803 15:55:14< mattsc> vultraz: let me just do a really quick test (with HttT)
20150803 15:57:52< mattsc> vultraz: http://imgur.com/AIFKJXq
20150803 15:58:01< mattsc> That’s what you meant, right?
20150803 15:58:27< vultraz> yeah. 
20150803 15:58:55< gfgtdf> mattsc:tey use teh same generator internalyl but have a idfferent interface
20150803 15:59:32< gfgtdf> Xudo: yes random_new::generator is autoamtically synced
20150803 15:59:33< mattsc> vultraz: I updated all the libraries at the time and that’s probably more than a year ago, but IIRC, it was the pango/cairo change that fixed that.
20150803 15:59:39< mattsc> gfgtdf: okay, thanks
20150803 15:59:46< mattsc> (I’ll be afk again for a while)
20150803 16:02:06-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150803 16:04:55-!- travis-ci [~travis-ci@ec2-54-92-247-35.compute-1.amazonaws.com] has joined #wesnoth-dev
20150803 16:04:56< travis-ci> wesnoth/wesnoth#6939 (master - d2375c0 : Charles Dang): The build passed.
20150803 16:04:56< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/73905200
20150803 16:04:56-!- travis-ci [~travis-ci@ec2-54-92-247-35.compute-1.amazonaws.com] has left #wesnoth-dev []
20150803 16:05:04< gfgtdf> 'as using the synced RNG would could OOS'
20150803 16:05:10< gfgtdf> this actually shouldnt be teh case
20150803 16:06:02< gfgtdf> the synced rng is smarth enough to notice that we are not in a synced event and thus just redirects to the unsynced random.
20150803 16:06:19< mattsc> gfgtdf: yes, it does.  the AI evaluation is only done on the host, and it is not done in replays.  And …
20150803 16:06:28< mattsc> gfgtdf: well, that was not the case some while ago
20150803 16:06:57< mattsc> where “some while ago” might mean 2 years, I don’t remember
20150803 16:07:05< gfgtdf> mattsc: that shoudl be the case since ~1.11.10
20150803 16:07:39< mattsc> gfgtdf: okay, I am pretty sure that I have always been using math.random since long before 1.11.10
20150803 16:07:51< mattsc> gfgtdf: so I learned something today  :)
20150803 16:10:19-!- travis-ci [~travis-ci@ec2-54-159-34-72.compute-1.amazonaws.com] has joined #wesnoth-dev
20150803 16:10:20< travis-ci> aquileia/wesnoth#2 (master - d2375c0 : Charles Dang): The build has errored.
20150803 16:10:21< travis-ci> Build details : https://travis-ci.org/aquileia/wesnoth/builds/73909821
20150803 16:10:21-!- travis-ci [~travis-ci@ec2-54-159-34-72.compute-1.amazonaws.com] has left #wesnoth-dev []
20150803 16:13:04-!- gfgtdf [~chatzilla@x50ab97f6.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 39.0/20150630154324]]
20150803 16:13:54-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 272 seconds]
20150803 16:14:18-!- horrowin1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has quit [Remote host closed the connection]
20150803 16:16:26-!- horrowind1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150803 16:22:49-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150803 16:25:36-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 255 seconds]
20150803 16:25:36-!- wedge010 is now known as wedge009
20150803 16:28:25-!- N4tr0n_ [~N4tr0n@cpe-76-175-70-130.socal.res.rr.com] has quit [Ping timeout: 250 seconds]
20150803 16:28:31-!- Kwandulin [~Miranda@p5B0096C3.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150803 16:31:02< aquileia> iceiceice, shadowm: According to https://github.com/travis-ci/apt-package-whitelist/blob/master/ubuntu-precise we could get all the packages travis needs via addons.apt.packages, which should a) get rid of the apt-get update timeouts b) allow us to use http://docs.travis-ci.com/user/migrating-from-legacy/
20150803 16:31:59-!- travis-ci [~travis-ci@ec2-54-87-112-73.compute-1.amazonaws.com] has joined #wesnoth-dev
20150803 16:32:00< travis-ci> gfgtdf/wesnoth-old#501 (lua_random - a795adc : gfgtdf): The build has errored.
20150803 16:32:00< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/73913277
20150803 16:32:00-!- travis-ci [~travis-ci@ec2-54-87-112-73.compute-1.amazonaws.com] has left #wesnoth-dev []
20150803 16:32:09< aquileia> Don't know whether that'd still be an issue if we get into the trusty beta, but can't hurt to mention it I guess
20150803 16:37:22< aquileia> wedge009: If you built some of wesnoth's GTK dependencies, would you mind sending them to me? Some of the ones in the external repo are quite outdated, and I wouldn't mind replacing them
20150803 16:38:18< aquileia> and if you compiled boost for VC2015, I could add a branch for that as well (VC2008 and VC2013 were built by me, VC2010 by gfgtdf)
20150803 16:44:15< Xudo> It will be very cool
20150803 16:46:34< Xudo> There those branches, exactly?
20150803 16:48:58< aquileia> You mean e.g. https://github.com/aquileia/external/tree/VC12 ?
20150803 16:49:40< aquileia> The only difference between the branches is the boost libs because they are compiler specific
20150803 16:50:46< quentinp> aquileia: in the meantime, I suggest using travis_wait on apt-get commands (http://docs.travis-ci.com/user/build-timeouts/#Build-times-out-because-no-output-was-received) I could send a PR
20150803 16:51:22< aquileia> So if I got VC2015 boost libs from e.g. wedge009 I could add a VC2015 branch
20150803 16:51:46< Xudo> I'll probably switch to MSVS2013 next weekend
20150803 16:53:02< aquileia> quentinp: I think using the travis apt addon should get rid of that timeout as well, but if your fix works, that'd be nice as well
20150803 16:53:21< aquileia> iceiceice is probably who you'd want to assign the PR to
20150803 17:00:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150803 17:04:10-!- cpf [~cpf@ip-188-118-61-46.reverse.destiny.be] has quit [Ping timeout: 240 seconds]
20150803 17:05:31-!- afre [63f401a5@gateway/web/freenode/ip.99.244.1.165] has joined #wesnoth-dev
20150803 17:12:05< mattsc> afre: hi.  You asked about AIs?
20150803 17:12:46< afre> Yes, which one in mp should I be testing, playing against?
20150803 17:14:30< mattsc> afre: well, the default AI is really the main AI out there.  Experimental AI is the default with a few additions (which, in principle, make it a bit better, but the changes are pretty subtle in most cases)
20150803 17:15:27< mattsc> afre: http://www.wesnoth.org/macro-reference.xhtml
20150803 17:15:40-!- GGeneral [~GGeneral@nat3.opti.net.ua] has joined #wesnoth-dev
20150803 17:15:57< mattsc> If you look for the EXPERIMENTAL_AI macro there, it tells you what is added
20150803 17:17:02< mattsc> So I would probably suggest using both the default and the experimental AI
20150803 17:18:02< mattsc> By contrast, “Fred”, which UnwiseOwl suggested is currently not in a playable state, and has never been good for playing against human players anyway.  I’m working on it again, but not sure if that’ll get anywhere.
20150803 17:18:42< afre> Is that "Strong AI" in 1.13.1?
20150803 17:20:05< mattsc> afre: don’t remember, but that name was very misleading anyway.  It was just the default AI with a very minor change to recruiting (which the person who redid the AI considered slightly stronger than the default).
20150803 17:20:23< mattsc> *who redid the recruiting code
20150803 17:22:34< afre> I see. Based on a few skimishes, Experimental AI seems to be less suicidal in 9p. Good job :)
20150803 17:22:50-!- GGeneral [~GGeneral@nat3.opti.net.ua] has quit [Remote host closed the connection]
20150803 17:24:49< mattsc> afre: well, good.  And thanks.  :)  As I said, I am still trying to get Fred to a state where he isn’t completely embarrassing (that’s on Freelands only), but quite honestly, I do not know any more if I’ll succeed with that.
20150803 17:24:57-!- chaverma [~Adium@209.63.146.244] has joined #wesnoth-dev
20150803 17:25:25-!- travis-ci [~travis-ci@ec2-54-87-112-73.compute-1.amazonaws.com] has joined #wesnoth-dev
20150803 17:25:26< travis-ci> gfgtdf/wesnoth-old#502 (lua_random - 4672f58 : gfgtdf): The build has errored.
20150803 17:25:26< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/73914723
20150803 17:25:26-!- travis-ci [~travis-ci@ec2-54-87-112-73.compute-1.amazonaws.com] has left #wesnoth-dev []
20150803 17:25:44-!- chaverma [~Adium@209.63.146.244] has quit [Client Quit]
20150803 17:28:55< aquileia> mattsc: I thought Fred was pretty effective before you refactored him?
20150803 17:29:46< aquileia> IIRC around 99% win ratio in a 1:1 agianst default AI?
20150803 17:31:10< mattsc> aquileia: yeah, against the default AI that was true, but he never did very well against human players
20150803 17:32:45< mattsc> I picked up working on the code again a while ago and am hoping to get somewhere with that eventually, but I have hoped that many times before  :P
20150803 17:33:02< aquileia> Sure, if you wanted him to win against Rigor...
20150803 17:33:06 * aquileia rolls eyes
20150803 17:33:08-!- N4tr0n_ [~N4tr0n@162.255.34.90] has joined #wesnoth-dev
20150803 17:33:37< mattsc> I’m not talking about winning.  Just about not doing really stupid things all the time.  :)
20150803 17:36:12-!- scorpion [~scorpion@46.166.188.218] has joined #wesnoth-dev
20150803 17:36:51< mattsc> aquileia: I am probably particularly negative right now because I spent a lot of effort onto making a number of “great changes”, and when I put it all together, man, does he play like crap!
20150803 17:37:16< mattsc> onto?  *on
20150803 17:37:46< mattsc> I still think that they are good changes overall, but that a lot more and a lot of finetuning is required.
20150803 17:39:23< aquileia> Well, the more aspects/CAs you add, the more parametres you have to fiddle with, I suppose
20150803 17:40:39-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150803 17:41:36< mattsc> Yep.  But my main problems right now are actually the larger strategic/tactical decisions: when to attack/hold/retreat; where and how to set up defensive lines, etc.
20150803 17:41:39-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has quit [Ping timeout: 250 seconds]
20150803 17:41:57< mattsc> And then translating that into evaluation equations that a computer can deal with.
20150803 17:42:27< mattsc> Anyways, I am working on it and hopefully I’ll be able to upload a revised (and maybe even better) version sometime 
20150803 17:53:01-!- irker584 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150803 18:03:38< afre> mattsc: (onto → in) ok, I will look into Default and Experimental. Fred isn't a selection so I don't have to worry about him atm.
20150803 18:05:34< mattsc> afre: sounds good (and thanks).  No, Fred is not (and has never been) in a state that he could be included in mainline, so he only exists in an add-on for now.
20150803 18:10:55< afre> I haven't seen Fred in ages. Maybe he moved :p
20150803 18:11:43-!- horrowind1 [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has quit [Remote host closed the connection]
20150803 18:11:43< mattsc> :)
20150803 18:14:00< mattsc> To those of you who don’t know the origin of the name:
20150803 18:14:33< mattsc> “It’s really too hard to say ‘advanced version of the Freelands grunt rush AI’ all the time.  We need a better name.”
20150803 18:14:37< mattsc> “How about Fred?"
20150803 18:16:58< pydsigner> Nice.
20150803 18:30:04< afre> Reminds me of Hugo the Abominable Snowman naming Daffy Duck "George".
20150803 18:39:38< midzer> i am trying to build for VC14. i get linker error for bzip2 and zlib
20150803 18:40:13< midzer> i built boost libs via http://stackoverflow.com/questions/7282645/how-to-build-boost-iostreams-with-gzip-and-bzip2-support-on-windows
20150803 18:41:35< aquileia> midzer: http://wiki.wesnoth.org/CompilingWesnothOnWindows#Generating_dependencies_for_other_versions
20150803 18:42:03-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20150803 18:42:57< aquileia> That's how I built the boost libs for VC9 and VC12 in the external package: https://github.com/aquileia/external/tree/VC12
20150803 18:44:38< aquileia> The instructions on the wesnoth wiki are enough, the additional commands listed in the repo are only to save disk space
20150803 18:45:59< aquileia> midzer: Also, wedge009 already built VC14 boost, but I don't have his files yet, so I couldn't include them into the repo
20150803 18:46:44-!- afre [63f401a5@gateway/web/freenode/ip.99.244.1.165] has quit [Quit: Page closed]
20150803 18:47:34< vultraz> you mean 15?
20150803 18:48:32< aquileia> VC 2014 = VC14
20150803 18:48:38< Xudo> compiler for MSVS2015 is vc14
20150803 18:48:42< aquileia> 2015
20150803 18:48:51< aquileia> what xudo said
20150803 18:49:19< Xudo> not sure if "compiler" is right word here though
20150803 18:50:34< vultraz> #confusing 
20150803 18:52:01< aquileia> welcome to Microsoft
20150803 18:52:18< Xudo> I stuck compiling for MSVS2015 at "unresolved external" error which says about boost::iostreams::zlib
20150803 18:52:42< Xudo> Other boost dependencies solved fine
20150803 18:54:19< aquileia> Xudo: Did you use an absolute path for -sZLIB_SOURCE= when compiling boost?
20150803 18:54:30< Xudo> yes i did
20150803 18:55:44< aquileia> And it created libboost_zlib-vc140-mt-1_58 ?
20150803 18:57:06< Xudo> yes
20150803 18:57:38< aquileia> Then I don't know what the issue might be...
20150803 18:58:35< Xudo> I'm not sure about exact error. let me run msvs...
20150803 18:58:36 * aquileia considers updating to VC14 just to try whether his instructions work there
20150803 19:03:03< Xudo> there was also an issue with snprintf constant
20150803 19:03:30< Xudo> I have fugured it out. may be it will save your time later
20150803 19:04:06< Xudo> in vc14 microsoft have defined snprintf by default and you need to add check for mscs version
20150803 19:04:15< Xudo> #if _MSC_VER<1900
20150803 19:04:19< Xudo>  #undef snprintf
20150803 19:04:25< Xudo>  #define snprintf _snprintf
20150803 19:04:26< Xudo> #endif
20150803 19:04:42< Xudo> global.hpp
20150803 19:06:15< Xudo> simple_wml.obj : error LNK2001: unresolved external symbol ""int const boost::iostreams::bzip2::ok"
20150803 19:06:28< Xudo> simple_wml.obj : error LNK2001: неразрешенный внешний символ ""int const boost::iostreams::zlib::best_speed"
20150803 19:07:05< aquileia> Xudo: src/SDL_gpu/SDL_gpu/GL_common/SDL_gpu_GL_matrix.c probably needs the same treatment...
20150803 19:07:13< Xudo> libboost_zlib-vc140-mt-1_58 is in include path
20150803 19:09:07< Xudo> bye. I need to go.
20150803 19:09:07< aquileia> Oh, and src/SDL_gpu/SDL_gpu/GL_common/SDL_gpu_GL_common.inl as well. Do you want to make a PR, or should I fix it?
20150803 19:09:53< Xudo> I'll try to make pull request. never did that before
20150803 19:10:01< Xudo> better to start from something simple)
20150803 19:10:03-!- ancestral [~ancestral@84.sub-70-197-199.myvzw.com] has joined #wesnoth-dev
20150803 19:10:09< Xudo> tomorrow..
20150803 19:10:20-!- Xudo [bce89833@gateway/web/freenode/ip.188.232.152.51] has quit [Quit: Page closed]
20150803 19:10:39< midzer> Xudo: my linker errors are gone for bzip2 and zlib. now just SDLmain.lib
20150803 19:11:17< midzer> i built boost libs again with exact commands via http://wiki.wesnoth.org/CompilingWesnothOnWindows#Generating_dependencies_for_other_versions
20150803 19:12:32< aquileia> midzer: Thanks for the heads up! Phew, now I don't have to update to VC14 to check if these instructions still work
20150803 19:17:39-!- ancestral [~ancestral@84.sub-70-197-199.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20150803 19:21:15-!- KinasisDrummer [~jason@90.204.17.155] has joined #wesnoth-dev
20150803 19:22:17-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150803 19:24:20-!- ancestral [~ancestral@63.92.240.233] has quit [Client Quit]
20150803 19:26:27-!- Appleman1234_ [~Appleman1@KD111107191033.au-net.ne.jp] has joined #wesnoth-dev
20150803 19:30:14-!- Appleman1234 [~Appleman1@KD036012009066.au-net.ne.jp] has quit [Ping timeout: 272 seconds]
20150803 19:35:29< aquileia> Xudo: You'll have to use #if defined(_MSC_VER) &&  _MSC_VER<1900 though - if undefined, _MSC_VER is assumed as 0 and thus < 1900 as well
20150803 19:37:48< midzer> \o/ successfully built
20150803 19:38:39< midzer> aquileia: Xudo is gone
20150803 19:39:39< aquileia> I know, but he might scan the logs for his name
20150803 19:40:31< aquileia> There's another option to make sure he is pinged...
20150803 19:40:36< aquileia> shikadibot: seen Xudo
20150803 19:40:37< shikadibot> aquileia: The person with the nick Xudo last spoke 30m 28s ago.  30m 17s ago they left with the message: Quit: Page closed
20150803 19:41:42-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150803 19:44:49-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150803 19:45:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150803 19:46:58-!- gfgtdf [~chatzilla@x50ab97f6.dyn.telefonica.de] has joined #wesnoth-dev
20150803 19:47:35< aquileia> shadowm: http://stackoverflow.com/a/2203694 - since we already depend on boost_filesystem, we might as well use it here for a platform independent working dir, I guess...
20150803 19:50:14< shadowm> aquileia: Have you seen the TS thread about UAC virtualization issues?
20150803 19:50:26< aquileia> no
20150803 19:50:36< shadowm> I had someone else stumble upon the same issue recently and I want to know exactly what the hell is going on.
20150803 19:52:59< aquileia> Do you mean http://forum.teamspeak.com/showthread.php/97184-Please-help-Had-to-turn-UAC-off-Lost-all-server-settings ?
20150803 19:53:17< shadowm> Uh.
20150803 19:53:22< shadowm> No. http://forums.wesnoth.org/viewtopic.php?f=4&t=42970
20150803 19:53:41< aquileia> Ah, technical support, not teamspeak...
20150803 19:53:45< pydsigner> He's not the only one who things "TeamSpeak" when you say TS
20150803 19:53:51< pydsigner> * thinks
20150803 20:01:11< aquileia> shadowm: I think we're both pretty sure what the issue is. I don't think I'll be able to fix it, though - I never really touched the filesystem code
20150803 20:01:42< shadowm> No, I'm not sure what the issue is, aquileia.
20150803 20:02:45< shadowm> I know we shouldn't be using this user-files-in-install-dir model we inherited from the XP times in the first place but it also seems before a few days ago people didn't run into problems like this.
20150803 20:04:08< aquileia> My guess: Our code tried to find the save location, but because the dir was 100% virtualized, it didn't recognize that the saves/ dir is where it should look. After he added the dir, the saves were found via virtualization
20150803 20:05:12< shadowm> Windows is supposed to make this transparent to us.
20150803 20:05:28< shadowm> We are not supposed to *see* where the virtualized paths really exists.
20150803 20:05:47< aquileia> So the issue is that the code didn't even check inside the dir (where it would have found the files) but only for its existence. Then, it fell back to a default I suppose, showing the contents of wesnoth/
20150803 20:05:52< shadowm> (Which is why there is no API to this mess.)
20150803 20:07:02< aquileia> As it only happened after a Windows update, I guess its a regression in finding directories...
20150803 20:11:39< aeth> Wesnoth doesn't seem to properly handle my right alt compose key, unlike every other application. ¡ shows up as !¡ because it's RightAlt ! ! and it reads both !s. Similarly, é shows up as 'é, etc.
20150803 20:12:10< aeth> This makes typing Spanish with a US keyboard really, really annoying because I have to delete extra characters
20150803 20:14:22< aeth> I don't recall having issues with this in other applications, even other games, even proprietary games that are terribly optimized for Linux.
20150803 20:16:25< aeth> Basically, it's reading dead keys for some reason
20150803 20:16:35< aquileia> aeth: Just a guess, but that might be one of the things the SDL2 transition will fix (Aginor is working on that)
20150803 20:16:48< aeth> aquileia: probably
20150803 20:17:17< aeth> aquileia: The SDL2 transition will introduce another really really annoying thing, though. It will minimize the Wesnoth window when I go to another window.
20150803 20:17:33< aeth> This has made many strategy games unplayable for me in my two monitors because I usually browse the web because they move slowly.
20150803 20:18:26< aquileia> aeth: You know of the movement acceleration setting, I hope?
20150803 20:19:15< aeth> In Wesnoth?
20150803 20:19:17< aeth> Of course I do.
20150803 20:19:19-!- genbattle [~genbattle@182.16.153.90] has joined #wesnoth-dev
20150803 20:19:24< aeth> What I mean is that it takes a very, very long time for people to move.
20150803 20:19:36< aquileia> Ah, you mean multiplayer
20150803 20:19:39< aeth> yes
20150803 20:19:50< aeth> It doesn't help that my RPGs are next-to-impossible :-)
20150803 20:20:14-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150803 20:22:04< aeth> Just testing with my SDL2 app, doesn't look like SDL2 makes it mandatory. So I guess it's just people on Steam somehow think it's a good idea when they update their engines.
20150803 20:23:19< aeth> Oh, it's only fullscreen mode. I see. This. http://www.ginkgobitter.org/sdl/?SDL_HINT_VIDEO_MINIMIZE_ON_FOCUS_LOSS
20150803 20:23:23< aeth> So it is a default.
20150803 20:23:55< aeth> Wesnoth iirc doesn't have multimonitor fullscreen so I have to just make it fullscreen manually at the moment. But when you do switch to SDL2 (which should make multi-monitor fullscreen trivial) please keep in mind to disable that flag
20150803 20:24:28< aeth> There's no reason for strategy games to be updated to do this. (e.g. now Paradox Interactive games do this and it's extremely annoying, especially because they sometimes crash when refocused)
20150803 20:25:37< aquileia> Aginor: ^
20150803 20:25:39< aeth> Testing it on my SDL2 app, and yes as soon as I make it fullscreen it uses this behavior.
20150803 20:26:07< aeth> This makes sense for e.g. an action game or an FPS but in a strategy game at least if you're doing serious MP, you will probably want to have a browser up because players take forever to think.
20150803 20:28:16< pydsigner> Very much so
20150803 20:28:22< aeth> (By "multi-monitor fullscreen" I mean "fullscreen on one monitor without making the other blank. Some games support it, some don't. My simple SDL2 app supports it so unless Wesnoth uses SDL poorly, it must be an SDL2 thing.)
20150803 20:28:40< pydsigner> I play with just a single monitor,
20150803 20:29:03< pydsigner> But I usually will Alt-Tab to do other stuff and wait for the desktop notification
20150803 20:29:16< pydsigner> Unless the game is actually moving quickly
20150803 20:29:19< aeth> Even back when I played Wesnoth on a laptop if I was in my room, I'd hook it into a 1920x1080 monitor and put the game there, and use the laptop monitor for e.g. the web
20150803 20:29:29< aeth> I guess it's less of a thing now that laptops have large resolutions
20150803 20:30:04< aeth> Also this loss of focus thing can cause bugs. Some games do crash when SDL2 does this, so it's probably not trivial to handle.
20150803 20:30:10< aeth> (At least in Linux.)
20150803 20:31:01< aquileia> aeth: As you alread worked with SDL2, the beast way to make sure these potential issues are handled correctly would be to offer Aginor your help ;)
20150803 20:31:39< aeth> I know the API, so I could probably help, but the SDL2 I've worked with is the Common Lisp bindings.
20150803 20:44:50-!- gfgtdf [~chatzilla@x50ab97f6.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 39.0/20150630154324]]
20150803 20:46:20< aquileia> loonycyborg: Using wesnoth_editor-icon for cwesnoth's start menu link was a copy paste error, I assume?
20150803 20:50:13< aquileia> loonycyborg: Also, would it be ok if the packaging process for the NSIS based updates would require a .7z or .tar.gz (or bzip2, although I'd have to look for that again) compression, and if yes, which would you prefer?
20150803 20:51:21< aquileia> That is, which format and which compression tool?
20150803 20:52:07< loonycyborg> aquileia: yes definitely copy paste error if it is
20150803 20:52:20< loonycyborg> no idea how tochoose compression
20150803 20:52:23< aquileia> ok, I'll fix it in the PR
20150803 20:53:24< aquileia> The compression question is only for a later PR, but I thought I'd ask now
20150803 20:55:37< aquileia> IF (and only if) we want to implement auto updates, we would have to bundle the files in an archive and upload it. NSIS has plugins for .zip, .7z, and .tar.gz
20150803 20:56:22< shadowm> 7z should provide the best compression of those three.
20150803 20:56:36< aquileia> .7z uses lzma and thus should bring the best compression, but I don't know whether you have a tool for that installed
20150803 20:57:31< aquileia> The NSIS plugin includes a minimalistic compression module, but you'd have to set it up so that scons finds it
20150803 20:58:27< midzer> what command parameter do i have to use when starting game in debug? there is an assert that game_config::path is empty
20150803 20:58:48< aquileia> loonycyborg: I don't suppose you already have 7zip installed (in hich case it would be a no-brainer) ?
20150803 20:59:42< shadowm> midzer: Add `--data-dir path/to/dist/dir` to the command line. The path should point to the directory that contains both data/ and src/.
20150803 21:00:28< loonycyborg> aquileia: I can install it
20150803 21:01:17< aquileia> midzer: You have to choose the debug configuration in MSVC to get full debug output. To debug your current build in MSVC, start it by pressing F5
20150803 21:01:19< midzer> game running. wohooo :D
20150803 21:02:13< midzer> i want to contribute some code with changes for those VC14 compilation, i think i do a PR
20150803 21:02:48-!- Appleman1234_ is now known as Appleman1234
20150803 21:02:49< aquileia> loonycyborg: Then I'll go for 7z
20150803 21:04:40< aquileia> Note that it'll take quite a while before it'll actually be relevant for you... I just nned to know with which of NSIS's plugins I need to work
20150803 21:04:49< aquileia> *need
20150803 21:14:14< aeth> iirc 7z is best compression but it's the slowest
20150803 21:14:44< aeth> so e.g. .7z or .tar.xz are usually used for things like backups
20150803 21:15:04< shadowm> We don't need speed here.
20150803 21:15:14< aeth> then go with that
20150803 21:15:55< aeth> .tar.gz is where you need to take some speed into account afaik and .tar.bz2 is largely irrelevant now that you can get 7z-like compression on Unix with .tar.xz
20150803 21:16:55< shadowm> xz is still slower than bzip2.
20150803 21:16:56< aeth> The problem with .7z is that it's designed for Windows. That's why there's https://en.wikipedia.org/wiki/Xz
20150803 21:16:57< midzer> does command line option "unit" for unit tests used to work?
20150803 21:17:17< aeth> shadowm: right bzip2 is in the middle of the tradeoff between speed (gz) and compression size (7zip/xz)
20150803 21:17:32< aeth> iirc
20150803 21:17:40< aeth> It's been a few years since I looked into the details of compression
20150803 21:17:50< shadowm> midzer: It still works since it's used by the WML unit tests pass done by Travis-CI as far as I know.
20150803 21:18:13< midzer> ah, its a headless unit test, without GUI running, i've seen wesnoth.exe been active
20150803 21:19:47< shadowm> If you are trying to run Wesnoth from a command prompt on Windows you might want to use the --wconsole switch to see the console output, although beware that if you are using cmd you should use a wrapper like cwesnoth.cmd instead unless you don't mind watching Wesnoth spit text while cmd.exe is accepting input.
20150803 21:20:53-!- Xara [~Yangyf@210.77.22.55] has joined #wesnoth-dev
20150803 21:21:27< aeth> iirc, though, using .7z on Linux or Unix is not a good idea. You'll lose information. Maybe even everything extracted will be marked as executable, not sure.
20150803 21:21:29< midzer> thanks shadowm
20150803 21:22:04< shadowm> aeth: You lose ownership and modes, yes.
20150803 21:23:04< shadowm> Nothing stops you from making a tar.7z, though.
20150803 21:23:27< aeth> It's probably not as efficient as a tar.xz I'm guessing?
20150803 21:23:34< aeth> There has to be a reason there's tar.xz
20150803 21:23:42< shadowm> People tend to overlook the fact that tar.gz/tar.bz2/tar.xz aren't formats either.
20150803 21:23:58< aeth> right, but xz is designed for use with tar
20150803 21:24:00< shadowm> You just make a tar archive and compress it.
20150803 21:24:39< aeth> right, iirc gz/bz2/xz are one-file compression. tar is the thing that makes it hold more than one
20150803 21:24:56< aeth> I'm guessing this is where the waste comes in when you do tar.7z
20150803 21:30:12< quentinp> shadowm: any opinion about the travis discussion we had earlier?
20150803 21:31:13< shadowm> quentinp: About asking to be added to the trusty beta?
20150803 21:31:52< quentinp> shadowm: yes, that and the OS X beta
20150803 21:33:05< shadowm> I don't have an opinion about the OS X beta because I'm not sure we have anyone able to set that up. As for trusty, yes, I plan to email them but I'd like to get the container-based setup working first.
20150803 21:33:56-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has joined #wesnoth-dev
20150803 21:35:07< quentinp> shadowm: it's easy to set up the OS X build and let it fail until it is fixed by someone else
20150803 21:35:21< quentinp> shadowm: if it does not get fixed, then after some time remove it from .travis.yml
20150803 21:35:39< quentinp> (let it fail = let this build fail without affecting the overall status)
20150803 22:01:29-!- joet [~joet@host86-163-217-190.range86-163.btcentralplus.com] has quit [Quit: Leaving]
20150803 22:03:01< aquileia> shadowm: For the container based setup, we'll need to issue quite a few PRs for https://github.com/travis-ci/apt-package-whitelist
20150803 22:03:38< aquileia> I was actually going through that list earlier today
20150803 22:05:25< aquileia> Or rather issues in their pre-defined format, which automatically generate PRs with build tests
20150803 22:06:28< shadowm> Yes.
20150803 22:07:38< aquileia> Ok, if you're working on that, I don't need to continue my own attempt
20150803 22:07:51< shadowm> I'm not actively working on it.
20150803 22:08:20< shadowm> You can see that the last time I tried was 22 days ago: https://travis-ci.org/shikadilord/wesnoth/jobs/70661388
20150803 22:08:34< shadowm> (Expand line 77.)
20150803 22:10:28< aquileia> We can circumvent the two libboost problems because they whitelisted other versions of them, but for all the others we need PRs (or the upgrade to SDL2)
20150803 22:10:46< aquileia> shadowm: I don't suppose you want us to wait for that?
20150803 22:11:22< shadowm> Of course not.
20150803 22:11:47< shadowm> 1.12 needs to be tested and we'll need to support it until 2018 at least.
20150803 22:15:27< aquileia> I wonder why you didn't need apt:   sources:   - ubuntu-toolchain-r-test   for gcc...
20150803 22:15:53-!- agorecki [~agorecki@unaffiliated/agorecki] has joined #wesnoth-dev
20150803 22:16:15< aquileia> Ah, you would've, that's why g++-4.8 didn't install
20150803 22:16:58< aquileia> shadowm: Should I take over with the container based builds? It would slow down progress on the Windows installer, though
20150803 22:17:12< shadowm> I had a brainfart and mistook saucy for precise, that's why I forgot to deal with gcc 4.8.
20150803 22:17:43-!- hay207 [~hay207@41.34.12.122] has quit [Ping timeout: 256 seconds]
20150803 22:17:58< shadowm> I think it's more important to focus on the Windows stuff for now so that we have something to show for 1.12.5.
20150803 22:18:13< aquileia> When is that going to be?
20150803 22:19:12< shadowm> As soon as we have substantial fixes in 1.12 mentioned in the changelog.
20150803 22:19:19< pydsigner> re 2018: So is 1.14 planned for 2016?
20150803 22:20:00< shadowm> As far as I can tell there's nothing important in 1.12 yet, but I could be wrong since gfgtdf appears to have forgotten to update the changelog again. 
20150803 22:20:20< shadowm> gfgtdf: d625d49d37969eb30ebe960c30a5d51b2ca1127e is not in the changelog and I want to know how important it is if at all.
20150803 22:21:48-!- KinasisDrummer [~jason@90.204.17.155] has quit [Quit: leaving]
20150803 22:21:57< shadowm> pydsigner: It's planned for 2039 right now, that may change soon.
20150803 22:24:03< pydsigner> TFW the next stable release is intentional vaporware
20150803 22:24:33< shadowm> Also, before deciding anything with regards to 1.12.5 I want to know what's going on in Windows land with the sudden UAC virtualization breakage.
20150803 22:32:15-!- matthew [~matthew@2602:30a:2e10:5c10:b9ac:bb0a:1eeb:f7dc] has joined #wesnoth-dev
20150803 22:32:19-!- Appleman1234_ [~Appleman1@KD111107185236.au-net.ne.jp] has joined #wesnoth-dev
20150803 22:32:39-!- matthew is now known as Guest89166
20150803 22:34:36-!- Appleman1234 [~Appleman1@KD111107191033.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20150803 22:41:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150803 22:43:15-!- gfgtdf [~chatzilla@x50ab96ce.dyn.telefonica.de] has joined #wesnoth-dev
20150803 22:43:45< gfgtdf> 20150803 22:20:20< shadowm> gfgtdf: d625d49d37969eb30ebe960c30a5d51b2ca1127e is not in the changelog and I want to know how important it is if at all.
20150803 22:45:06< gfgtdf> shadowm: mostly if fixes "enemies defeated" in some situations in mp and alst when scenario ends in networked mp
20150803 22:45:59< gfgtdf> shadowm: specially it fixes a bug reported here: http://forums.wesnoth.org/viewtopic.php?f=8&t=32384&start=4530#p587601
20150803 22:47:10< gfgtdf> shadowm: more importantly than this bug, i also fix some parts of LoW in 1.12
20150803 22:47:16< shadowm> Okay, sounds important. Make sure to add this to the changelog.
20150803 22:48:45-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150803 22:52:09< aquileia> gfgtdf: Just FYI, the main part of an auto updater NSIS script should be as simple as http://pastebin.com/k50St8ba
20150803 22:52:47< aquileia> Most of the work is parsing and correcting the registry and shortcuts, the sctual update is trivial
20150803 22:52:57< aquileia> *actual
20150803 22:53:07< gfgtdf> aquileia: why do you want to correct the shortcuts ?
20150803 22:53:18< gfgtdf> aquileia: i think moving teh exe around when updating is a bad idea
20150803 22:53:23< gfgtdf> aquileia: the*
20150803 22:53:32< aquileia> Because they are in a folder called BfW 1.12.4
20150803 22:53:44< gfgtdf> aquileia: also note that the user might have made some shortcuts themself
20150803 22:54:06< gfgtdf> aquileia: i suggest to remove teh folder to 1.14 for 1.14 version
20150803 22:54:19-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has joined #wesnoth-dev
20150803 22:54:28< gfgtdf> aquileia: dropping teh throd numver from teh path i mean as sson as we have an auto updater
20150803 22:54:31< gfgtdf> third
20150803 22:54:59< aquileia> We might need versioned folders, else people couldn't install multiple patch versions
20150803 22:55:33< gfgtdf> aquileia: if they really want differnt versions, they can still rename teh folder manually or install into different location manually
20150803 22:55:36< aquileia> Not that anyone using a Windows installer for Wesnoth would do that
20150803 22:56:40< gfgtdf> aquileia: i think it easy enough
20150803 22:56:41< aquileia> gfgtdf: Changing the install location is a simple "Rename $INSTDIR $0" as soon as we have parsed the registry
20150803 22:57:08< gfgtdf> aquileia: but we migth break the users shortcuts that he mademanually
20150803 22:57:32< gfgtdf> aquileia: for example one might have made a desktop shortcut whcih will then be broken
20150803 22:57:57< gfgtdf> aquileia: or a shortcut to startit as debug mode
20150803 22:58:44< aquileia> gfgtdf: Would you write the C++ side code (somehow find out that wesnoth-x was released, call WUpdate.exe -x from inside wesnoth.exe) ?
20150803 22:59:51< aquileia> gfgtdf: We can make 'Change the install location to BfW 1.12.5?' an optional feature
20150803 22:59:52< UnwiseOwl> This is only for stable rek
20150803 23:00:09< UnwiseOwl> Releases, right?
20150803 23:01:11< aquileia> UnwiseOwl: Stable probably won't get the auto updater, but it'll get update packages (download, execute, voila)
20150803 23:02:02< UnwiseOwl> But automaticalt
20150803 23:02:03< gfgtdf> aquileia: you mean now for the 'manual' installer which has to be downloaded or the auto updater ?
20150803 23:02:07< aquileia> gfgtdf: ./WUpdate.exe, to be specific
20150803 23:02:41< UnwiseOwl> Hrmm, I'll come back later
20150803 23:03:14< shadowm> If the auto updater functionality requires changes to Wesnoth's source, I'm willing to evaluate backporting them with the usability bug excuse.
20150803 23:03:19< aquileia> gfgtdf: I'm working on the manual updater, but if you provide the C++ side call to an auto updater, I think the auto updater is pretty simple (only problem remaining would be where to host the patch files)
20150803 23:03:42< aquileia> shadowm: Ok, that's good to hear
20150803 23:05:09< gfgtdf> aquileia: Calling WUpdate.exe is easy i'd say, but i dont eally know how to write the client/server beast
20150803 23:06:26< aquileia> gfgtdf: That's why I'd suppose a hack - we could use e.g. the MP lobby chat to get that information
20150803 23:07:33< gfgtdf> aquileia: you man liek pasrng teh mp lobby welcome message?
20150803 23:07:43< aquileia> Exactly
20150803 23:08:28< gfgtdf> aquileia: im not sure that this also gives informstion about sub versions liek 1.12.5b, i know we sonethimes have such versions and somethimes onyl for windows.
20150803 23:09:42< aquileia> We could add that information to the message and hack the message display not to show it
20150803 23:09:50< aquileia> Just an option
20150803 23:10:33< gfgtdf> aquileia: hm thats quite hacky.
20150803 23:10:52< aquileia> I didn't say we should, but we 'could'
20150803 23:12:22< gfgtdf> aquileia: hm
20150803 23:13:18-!- irker430 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150803 23:13:18< irker430> wesnoth: gfgtdf wesnoth:1.12 2ca6169c7ac2 / changelog: Update changelog http://git.io/vO0FR
20150803 23:14:01< aquileia> And who knows, perhaps the lobby code is designed in a way that we actually only need to call one of its functions without actually messing with the lobby code
20150803 23:15:15< aquileia> Then we add the user "Wesnoth version announcement" to everyone's ignore list so that they don't see the messages, and use that as our information channel
20150803 23:15:22< gfgtdf> aquileia: im currently looking at the campaignd code to check how had it woudl be to make  asepereate 'tell me teh current version' server
20150803 23:16:31< aquileia> gfgtdf: If you don't mind I'd go to bed, it's pretty late (I'll read the logs)
20150803 23:16:40< gfgtdf> aquileia: ok
20150803 23:17:06-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has quit [Quit: Page closed]
20150803 23:30:26< aeth> I'm just curious... why 1.14 and not 2.0?
20150803 23:30:51< aeth> The game's a pretty different game from 1.2 already, but if it's updated to SDL 2 and a few more significant changes get in, that's *huge* and imo justifiable for 2.0
20150803 23:31:00< shadowm> Because I get to decide this now and I've decided there won't be a 2.0.
20150803 23:31:05< shadowm> :)
20150803 23:31:20< aeth> Wesnoth's version system copied the Linux kernel iirc. Except the Linux kernel gave that up years ago, in 2011 or so.
20150803 23:31:25< aeth> And now they do inflated numbers like everyone else.
20150803 23:31:35< shadowm> No, they abandoned it with 2.6.x.
20150803 23:32:04< aeth> shadowm: well 2.6.x was essentially equivalent to (2.6).x.x if you wanted to parse it
20150803 23:32:11< aeth> iirc
20150803 23:32:27< shadowm> But it did away with the dev/stable branch distinction.
20150803 23:33:06< aeth> shadowm: There are very few things that would justify a 2.0, and that would include SDL 2.0 and what it could offer, as well as e.g. substantial multiplayer changes like allowing simultaneous same-team turns.
20150803 23:33:07< shadowm> I.e. 2.2 vs. 2.3 and 2.4 vs. 2.5. There wasn't a 2.7 dev branch.
20150803 23:33:44< aeth> If the move to SDL 2.0 was 2.0 then the real major changes would come in 2.2 and 2.4, with e.g. fancy shader effects for certain things
20150803 23:33:45< shadowm> aeth: Just know that whether 1.14 will get to be 2.0 or not isn't up for debate.
20150803 23:34:05-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20150803 23:36:04< shadowm> I really hate to be curt like that, but I have read countless variations of the same arguments time and again even before I took over; I don't need to be lectured on the matter.
20150803 23:37:01< aeth> I'm just saying that the way software version numbers work is very different than ten years ago.
20150803 23:37:03< UnwiseOwl> Admit it, you LOVE being curt like that :p
20150803 23:37:06< aeth> Version numbers are essentially a convention.
20150803 23:37:14< aeth> And when you don't keep up with the conventions you seem archaic.
20150803 23:37:28< shadowm> The way software version numbers work is the way each project's release team decides, aeth.
20150803 23:38:11< shadowm> Just go and ask Mozilla and Chromium and Canonical and Microsoft and KDE and Linux and Apple.
20150803 23:38:30< aeth> Mozilla and Linux both changed the way they handled version numbers since you started being involved in Wesnoth, shadowm.
20150803 23:38:45-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 256 seconds]
20150803 23:38:45< shadowm> And both do it differently, aeth.
20150803 23:40:04< shadowm> Since 3.0, Linus every once in a while bumps the major number for no particular reason at all (he's admitted there's no actual logic to it). There's nothing spectacular whatsoever that makes Linux 4.0 different from 3.19 or 3.18.
20150803 23:40:24< shadowm> Not in a way that would contrast with the differences bewteen 3.19 and 3.18 or 3.18 and 3.17, etc.
20150803 23:40:39< shadowm> Mozilla bumps the major number for every single feature release, Linus doesn't.
20150803 23:40:56< aeth> Yes, the Mozilla strategy would be sort of like Wesnoth just dropping the 1.
20150803 23:41:08< aeth> In fact, iirc some software skipped numbers to keep up with version number inflation
20150803 23:41:12< shadowm> Canonical uses the release date as their version number.
20150803 23:41:38< aeth> Canonical's version numbering system is odd. Microsoft's irrelevant (this is about open source software).
20150803 23:41:40< shadowm> Apple follows a linear progression but hasn't bumped the major number ever since the great shift from Mac OS 9 to Mac OS X.
20150803 23:41:47< shadowm> Which was really, really great.
20150803 23:42:19< shadowm> No, this is not exclusively about open source software.
20150803 23:42:48< shadowm> But if you want more OSS examples, look at systemd.
20150803 23:43:04< shadowm> Then look at GNU Emacs.
20150803 23:43:14< shadowm> Then GNU libc.
20150803 23:43:19< shadowm> GCC.
20150803 23:43:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150803 23:43:39< shadowm> Et cetera. You'll find that there isn't a single convention that's respected by everybody.
20150803 23:44:10< aeth> There are general patterns and trends.
20150803 23:44:12< shadowm> We get to decide our own convention, and I say our existing convention is good for my term.
20150803 23:44:32< aeth> And, anyway, I don't think you understand how transformative of an engine upgrade SDL2 could offer.
20150803 23:44:34< shadowm> And that's all there's to say on the matter, really, unless you want to keep wasting both my time and yours.
20150803 23:46:36-!- UnwiseOwl [~UnwiseOwl@1.124.48.176] has quit [Read error: Connection reset by peer]
--- Log closed Tue Aug 04 00:00:12 2015