--- Log opened Thu Sep 15 00:00:00 2016 20160915 00:08:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 00:08:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 00:09:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 00:09:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 00:10:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 00:23:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 00:33:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 00:33:52-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Read error: Connection reset by peer] 20160915 00:39:21-!- ancestral [~ancestral@245.sub-174-219-0.myvzw.com] has joined #wesnoth-dev 20160915 00:53:34-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 00:58:16-!- ancestral [~ancestral@245.sub-174-219-0.myvzw.com] has quit [Quit: i go nstuf kthxbai] 20160915 00:59:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20160915 01:05:07-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 01:09:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds] 20160915 01:12:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Ping timeout: 240 seconds] 20160915 01:13:18-!- gfgtdf [~chatzilla@x4e368107.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617]] 20160915 01:17:04-!- Bonobo [~Bonobo@2001:44b8:254:3200:a0bb:5dc3:f5e7:623b] has joined #wesnoth-dev 20160915 01:17:11-!- Bonobo [~Bonobo@2001:44b8:254:3200:a0bb:5dc3:f5e7:623b] has quit [Client Quit] 20160915 01:17:25-!- Bonobo [~Bonobo@2001:44b8:254:3200:a0bb:5dc3:f5e7:623b] has joined #wesnoth-dev 20160915 01:24:17-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 01:26:06-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20160915 01:27:39< vultraz> celmin: http://en.cppreference.com/w/cpp/io/basic_istringstream/swap2 20160915 01:27:45< vultraz> celmin: the commented out thing in my commit was the displaying of credits in the help browser 20160915 01:27:50< vultraz> need to decide if i want to keep it 20160915 01:28:13< vultraz> gfgtdf: IIRC we made maps take no borders and masks use border=yes/no to decide whether to give it a border of 1/0. 20160915 01:28:22< vultraz> border_size isn't used anymore I think 20160915 01:29:58-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has joined #wesnoth-dev 20160915 01:36:30< celmin> Hi ancestral 20160915 01:39:15-!- Bonobo [~Bonobo@2001:44b8:254:3200:a0bb:5dc3:f5e7:623b] has quit [Ping timeout: 248 seconds] 20160915 01:42:24< vultraz> celmin: do *you* think we need to keep credits in help? 20160915 01:42:29< vultraz> it's rather redundant 20160915 01:49:11-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed] 20160915 01:49:54< celmin> Dunno. 20160915 01:50:05< celmin> Doesn't hurt. 20160915 01:50:35< celmin> It does mean they can be viewed eg from in-game or from the lobby. 20160915 01:54:38-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20160915 01:56:56-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20160915 01:58:21< mattsc> Well, I downgraded back to Xcode 7.3.1. 20160915 01:58:29< mattsc> Unfortunately, that did not fix the problem. 20160915 01:59:00< celmin> Maybe you need to downgrade the dev tools that are installed into /usr etc? 20160915 02:00:43< mattsc> Did that too ... 20160915 02:01:04< mattsc> However, I just removed the Wesnoth 1.13 data dir and guess what ... 20160915 02:01:32< mattsc> Somebody hand be a 2x4 please… 20160915 02:01:39< mattsc> s/be/me 20160915 02:03:09< celmin> Eh? 20160915 02:03:38< celmin> I feel like this implies it could be something about the WML cache... 20160915 02:03:53< celmin> Which seems to have been broken lately... 20160915 02:03:58< mattsc> No, I had already removed the cache previously, that’s not where the problem was. 20160915 02:04:02< celmin> Ah. 20160915 02:04:26< celmin> By data dir, you mean in Application Support? 20160915 02:05:32< mattsc> yes 20160915 02:05:39< mattsc> save_index is the culprit 20160915 02:05:51< celmin> I see... 20160915 02:06:15< mattsc> I’ve been switching things back and forth between the old and the new … I guess it’s user dir not data dir 20160915 02:07:06< mattsc> Blargh. I spent hours trying to hunt this down, downgraded Xcode, and all for this. Grmpf. 20160915 02:08:18< mattsc> I guess I should be happy that it was only hours. 20160915 02:12:11-!- Bonobo [~Bonobo@129.127.113.165] has joined #wesnoth-dev 20160915 02:19:11< celmin> Well, I'm glad you solved it at least... 20160915 02:21:44< mattsc> Yeah, so am I. Thanks. 20160915 02:28:00-!- Bonobo [~Bonobo@129.127.113.165] has quit [Ping timeout: 260 seconds] 20160915 02:31:44-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 265 seconds] 20160915 03:10:39-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev 20160915 03:10:40-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Changing host] 20160915 03:10:40-!- iceiceice [~chris@unaffiliated/iceiceice] has joined #wesnoth-dev 20160915 03:14:05< shadowm> src/tests/test_filesystem.cpp(70): error: in "filesystem/test_fs_base": check is_relative("/foo/..") has failed 20160915 03:14:26< shadowm> What. Ever. 20160915 03:17:30< shadowm> Oh well I'll probably have to upload like 90 MiB to my fork. I can't be bothered to go out of my way to save up bytes this time. 20160915 03:20:50< shadowm> Well, that didn't go well at all. 20160915 03:21:51< shadowm> I'll just delete and refork as usual. 20160915 03:22:51-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!] 20160915 03:23:17< shadowm> Finally. God. 20160915 03:27:43-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] 20160915 03:28:54-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20160915 03:29:56< shadowm> vultraz: PR #778. 20160915 03:30:27< vultraz> noted 20160915 03:31:56< shadowm> Do you still use CodeBlocks? 20160915 03:32:31-!- fabi [~fabi@176.4.88.150] has joined #wesnoth-dev 20160915 03:33:19< shadowm> I went and did the commit anyway. 20160915 03:33:41< shadowm> Can't be bothered to meld it into the first one right now. 20160915 03:33:58< vultraz> yes I use CB 20160915 03:36:28< Aginor> vultraz: if you're around later I should finally have some time 20160915 03:36:37< shadowm> Everyone else: feel free to review my code and tell me how much it sucks. 20160915 03:36:44< vultraz> Aginor: will be 20160915 03:37:13< Aginor> sweet 20160915 03:37:26< Aginor> probably 3-4h away for me 20160915 03:46:31-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 03:48:38-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009] 20160915 03:48:43-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Client Quit] 20160915 03:48:59-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev 20160915 03:49:38< Aginor> https://github.com/blog/2256-a-whole-new-github-universe-announcing-new-tools-forums-and-features 20160915 03:49:43< Aginor> that's a bunch of nice stuff 20160915 03:59:32-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20160915 03:59:38-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20160915 04:33:57-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20160915 04:42:48-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Read error: Connection reset by peer] 20160915 04:42:53-!- Jetrel_ [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev 20160915 04:45:03-!- Bonobo [~Bonobo@2001:44b8:254:3200:7938:fdff:5342:be75] has joined #wesnoth-dev 20160915 04:49:55-!- Bonobo [~Bonobo@2001:44b8:254:3200:7938:fdff:5342:be75] has quit [Ping timeout: 265 seconds] 20160915 04:50:15-!- Bonobo [~Bonobo@2001:44b8:254:3200:5c9:df2:e348:aa8f] has joined #wesnoth-dev 20160915 04:56:56-!- irker112 [~irker@uruz.ai0867.net] has joined #wesnoth-dev 20160915 04:56:57< irker112> wesnoth: Charles Dang wesnoth:master 1fb26d9e1c01 / / (48 files in 14 dirs): Convert remaining SDL1.2 defines to their SDL2 counterparts and dropped compat l https://github.com/wesnoth/wesnoth/commit/1fb26d9e1c01cca0753bdb905ded29bad6d69757 20160915 05:00:45< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup 633d7d1c0394 / src/ (21 files in 5 dirs): Cleaned up handling of basic alpha setting https://github.com/wesnoth/wesnoth/commit/633d7d1c039436959ce04544c8fa263ce5f57ae4 20160915 05:00:48< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup da818c186750 / src/sdl/utils.cpp: Removed old Pandora compatibility code https://github.com/wesnoth/wesnoth/commit/da818c186750014a2fc32ed7b6b4a563696f365a 20160915 05:00:51< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup ed8c8fbd396e / src/ (5 files in 2 dirs): Refactored out create_optimized_surface https://github.com/wesnoth/wesnoth/commit/ed8c8fbd396eb0a9643a6993ea9d550033b35adb 20160915 05:00:55< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup 646ee1dc68a6 / src/sdl/utils.hpp: Refactored use of util::scoped_resource out of surface struct (now class) https://github.com/wesnoth/wesnoth/commit/646ee1dc68a6b9f439f424160db1832143441a8f 20160915 05:00:57< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup 75eda82c9932 / src/ (gui/core/canvas.cpp sdl/utils.hpp): Allow any surface to create its own SDL software (surface) renderer https://github.com/wesnoth/wesnoth/commit/75eda82c9932fda795e0c958160e944edc6b2c4c 20160915 05:01:00< irker112> wesnoth: Charles Dang wesnoth:surface_cleanup bdf57214e834 / / (6 files in 4 dirs): Split drawing primitives helpers out of the GUI2 canvas and into their own file https://github.com/wesnoth/wesnoth/commit/bdf57214e834ce6f10c1525b524694d02ac5189e 20160915 05:16:28-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 05:16:29< travis-ci> wesnoth/wesnoth#10930 (master - 1fb26d9 : Charles Dang): The build is still failing. 20160915 05:16:29< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160072733 20160915 05:16:29-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 05:19:10< vultraz> can i just tell travis to ignore that line :| 20160915 05:22:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 05:26:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds] 20160915 05:28:37< irker112> wesnoth: Charles Dang wesnoth:master e78777caa391 / src/ (10 files in 4 dirs): Replaced instances of util::scoped_ptr with std::unique_ptr https://github.com/wesnoth/wesnoth/commit/e78777caa391f14ab1cf5a8d13996b6d81533296 20160915 05:28:39-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 05:28:40< travis-ci> wesnoth/wesnoth#10931 (surface_cleanup - bdf5721 : Charles Dang): The build was broken. 20160915 05:28:40< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160073024 20160915 05:28:40-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 05:35:11-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Remote host closed the connection] 20160915 05:36:37-!- Kwandulin [~Miranda@p200300760F2C71F8C1B8DFCCD1DCB168.dip0.t-ipconnect.de] has joined #wesnoth-dev 20160915 05:39:39-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai] 20160915 05:52:32-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev 20160915 06:05:08-!- atarocch [~atarocch@HSI-KBW-46-223-95-173.hsi.kabel-badenwuerttemberg.de] has joined #wesnoth-dev 20160915 06:07:45-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 06:07:46< travis-ci> wesnoth/wesnoth#10933 (master - e78777c : Charles Dang): The build is still failing. 20160915 06:07:46< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160075370 20160915 06:07:46-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 06:18:14-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 06:21:00-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Client Quit] 20160915 06:28:28< irker112> wesnoth: Charles Dang wesnoth:master 3eac7281d936 / src/ (scoped_resource.hpp tools/exploder_utils.cpp): Convert uses of util::scoped_array and util::scoped_file to std::unique_ptrs https://github.com/wesnoth/wesnoth/commit/3eac7281d936987fc722fa2943ecf6ae11386703 20160915 06:28:31< irker112> wesnoth: Charles Dang wesnoth:master 04d012a6923d / src/ (generators/map_create.cpp scoped_resource.hpp): Removed util::scoped_ptr, util::scoped_array and util::scoped_file code https://github.com/wesnoth/wesnoth/commit/04d012a6923de41a0149972e2d6a2687a105f9a7 20160915 06:28:53< vultraz> alright, once the campaignd and my surface_cleanup branches are merged we can delete scoped_resource.hpp entirely 20160915 06:47:36< irker112> wesnoth: Charles Dang wesnoth:master c99998cb521d / src/about.cpp: Minor cleanup https://github.com/wesnoth/wesnoth/commit/c99998cb521d1a71de76e17a0d0b152c9e630cf9 20160915 06:52:54-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 06:52:55< travis-ci> wesnoth/wesnoth#10936 (master - 04d012a : Charles Dang): The build is still failing. 20160915 06:52:55< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160082458 20160915 06:52:55-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 06:52:59-!- boucman_work [~boucman@193.56.60.161] has joined #wesnoth-dev 20160915 06:55:01-!- Kwandulin [~Miranda@p200300760F2C71F8C1B8DFCCD1DCB168.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds] 20160915 07:01:45-!- Kwandulin [~Miranda@p200300760F2C71D3C1B8DFCCD1DCB168.dip0.t-ipconnect.de] has joined #wesnoth-dev 20160915 07:06:47-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 07:06:48< travis-ci> wesnoth/wesnoth#10937 (master - c99998c : Charles Dang): The build is still failing. 20160915 07:06:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160084648 20160915 07:06:49-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 07:10:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 07:11:25-!- atarocch [~atarocch@HSI-KBW-46-223-95-173.hsi.kabel-badenwuerttemberg.de] has quit [Ping timeout: 244 seconds] 20160915 07:11:32< irker112> wesnoth: Charles Dang wesnoth:master 5004f9bfaa33 / src/gui/dialogs/end_credits.cpp: Attempt to appease travis https://github.com/wesnoth/wesnoth/commit/5004f9bfaa33f666f381224de642128ff66be6a2 20160915 07:14:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds] 20160915 07:20:34-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20160915 07:22:42-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind] 20160915 07:25:59-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev 20160915 07:30:56-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 07:30:57< travis-ci> wesnoth/wesnoth#10938 (master - 5004f9b : Charles Dang): The build is still failing. 20160915 07:30:58< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160088436 20160915 07:30:59-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 07:34:31< zookeeper> vultraz, yes credits should to be kept in help because otherwise there's no convenient way to view them in-game. the credits screen is obviously not a convenient way. 20160915 07:34:41< zookeeper> (unless you make it scrollable etc) 20160915 07:45:44< Aginor> hey everyone 20160915 07:49:18< zookeeper> _o/ 20160915 07:53:57< vultraz> zookeeper: it is scrollable now but why in hell would you want to browse the *credits* in-game? 20160915 07:54:22< Aginor> vultraz: how can I help? 20160915 07:54:26< zookeeper> for the same reason anyone ever looks at credits..? 20160915 07:55:08< vultraz> zookeeper: most people would look at the credits maybe once, and most certainly not in-game. 20160915 07:55:29< vultraz> Aginor: I wanted to discuss a few design details 20160915 07:55:38< zookeeper> vultraz, insufficient data for conclusion 20160915 07:55:54< shadowm> Most people don't look at the unit descriptions either. 20160915 07:56:00< shadowm> Let's remove all unit descriptions. 20160915 07:56:03< zookeeper> i see no reason to remove them from the help 20160915 07:56:06< shadowm> /s 20160915 07:56:09< zookeeper> it's not like they're in the way or anything 20160915 07:56:21< vultraz> Not to mention the credits have always produced *major* lag when viewing them in help. 20160915 07:56:47< Aginor> hmm 20160915 07:57:03< Aginor> I'm getting an selinux violation when building wesnoth now 20160915 07:57:15< vultraz> i can restore them, sure, but I'd like to know why you say people should want to see them in-game 20160915 07:57:26< shadowm> Naughty compiler. 20160915 07:57:36< Aginor> autorevision 20160915 07:58:07< vultraz> anyway.. 20160915 07:58:16< vultraz> Aginor: have you had a chance to take a look at my PR? 20160915 07:58:18< shadowm> vultraz: The big question here is why you need to remove every single thing you come across. 20160915 07:58:32< zookeeper> vultraz, well, i mean if the credits screen is now sufficient for viewing them then they don't need to be kept in the help i think. but i don't know if it is, because i've not seen and tried it. 20160915 07:58:38< zookeeper> and yes, ^ that too 20160915 07:58:43< Aginor> vultraz: I've glanced at it briefly, but I haven't had much time or energy lately 20160915 07:58:58< vultraz> shadowm: I disabled it in help during my refactor. I'm pondering if it's worth replacing. 20160915 08:00:44< vultraz> Aginor: alright, well, the general thing is I added the ability to create a software renderer context for any surface. 20160915 08:01:24< vultraz> My idea was to use a general system for drawing primitives and replace stuff like fill_rect, fill_rect_alpha, and draw_solid_tinited_rectangle 20160915 08:01:28< vultraz> but.. 20160915 08:01:38< vultraz> that doesn't.. exactly work 20160915 08:02:03< vultraz> if I want to use renderer code I need a renderer context, and yes, it can be created, but it's rather messy 20160915 08:02:12< vultraz> and likely not very performance-friendly. 20160915 08:02:21< Aginor> yeah... I can see a few things with the overall idea that's a bit unideal... 20160915 08:02:50< vultraz> but I also don't like keeping multiple functions around that do similar things. 20160915 08:03:28< Aginor> what's the purpose of making a rendering framework that renders onto an in-memory bitmap? 20160915 08:03:42< Aginor> what are you trying to achieve? 20160915 08:03:44-!- Bonobo [~Bonobo@2001:44b8:254:3200:5c9:df2:e348:aa8f] has quit [Ping timeout: 265 seconds] 20160915 08:04:04-!- Bonobo [~Bonobo@2001:44b8:254:3200:e51f:7dcf:bc76:9e6e] has joined #wesnoth-dev 20160915 08:05:06< vultraz> Well, because of spixi's chat widget prototype on the forums i realized one could use SDL functions to draw simple shapes like lines, etc, instead of relying on custom implementations like the canvas did. 20160915 08:05:27< vultraz> As you know, I implemented that in the canvas and noticed an improvement in performance, though I cannot quantify that. 20160915 08:06:21< vultraz> So I figured it'd be more performance-friendly to reduce the number of direct surface manipulations done in general. 20160915 08:06:40< Aginor> but you ran into the alpha handling? 20160915 08:06:49< vultraz> Which is also why I did some refactoring to slightly reduce the number of intermediate surfaces created. 20160915 08:07:23< vultraz> Aginor: no, I only ran into problems with alpha when I tried to make canvas *images* use textures and SDL_RenderCopy. 20160915 08:07:48< vultraz> (which I then abandoned) 20160915 08:07:53< vultraz> I have done some cleanup on surface alpha manipulation in my branch, but that is unrelated. 20160915 08:08:13-!- fabi [~fabi@176.4.88.150] has quit [Read error: Connection reset by peer] 20160915 08:09:22< vultraz> the general state of affairs is a large number of intermediate surface objects get created and manipulated or blitted when doing anything in general. 20160915 08:09:48< vultraz> though whether that actually copies the SDL_Surface objects since pointers are passed around, I'm not sure 20160915 08:09:54< vultraz> perhaps it's less performance-hungry than I though 20160915 08:10:04< vultraz> still, it'd be better to avoid always blitting 20160915 08:10:34< Aginor> most of it, except the create functions, passes pointers around 20160915 08:10:45< vultraz> not to mention we have 2 forms of blitting - SDL_BlitSurface and our custom implementation of blitting :/ 20160915 08:11:04< Aginor> the sdl_blitsurface is the most common one 20160915 08:11:10< Aginor> the other one needs to be removed 20160915 08:11:19< Aginor> but just replacing it doesn't work :/ 20160915 08:11:30< vultraz> right 20160915 08:11:33< vultraz> for example 20160915 08:11:44< vultraz> I realized text and images in gui2 use the custom blit function 20160915 08:11:51< vultraz> i tested both the sdl_blitsurface 20160915 08:11:55< vultraz> s/the/with 20160915 08:12:00< vultraz> images work fine 20160915 08:12:07< vultraz> but certain images have slightly different alpha 20160915 08:12:17< vultraz> and i don't know which is the "right" one :/ 20160915 08:12:32< Aginor> then we need to fix gui2 20160915 08:12:42< Aginor> and what exactly do you mean by "right" alpha? 20160915 08:12:55< vultraz> well, which is the "correct" rendering? 20160915 08:13:04< vultraz> ie, true-to-image. 20160915 08:13:07< Aginor> what do you mean? 20160915 08:13:43< Aginor> sdl_blitsurface I'd say 20160915 08:14:16< vultraz> let me get a screenshot 20160915 08:15:24< vultraz> current master, gui2 canvas, custom blit: https://drive.google.com/file/d/0B-mR9s8FduLLcFpNN1ZmQ0twaVE/view?usp=sharing 20160915 08:16:08< vultraz> same thing, but using sdl_blitsurface https://drive.google.com/file/d/0B-mR9s8FduLLR2VXSmNScnZMLW8/view?usp=sharing 20160915 08:16:25-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 08:17:23< JyrkiVesterinen> vultraz: The difference between screenshots looks like one implementation assumes premultiplied alpha while the other doesn't. 20160915 08:18:40< tad_> zookeeper: Do you know what "modify 20160915 08:19:06< tad_> zookeeper: Do you know what "modify_placing" means .. I think it's a [scenario] attribute. 20160915 08:19:40< zookeeper> sounds like that thing that relates to castle/leader positions in random maps 20160915 08:21:08< zookeeper> probably exactly what the code comment says 20160915 08:21:42< vultraz> Aginor: the bottom line is i want to reduce the number of times things are blitted or drawn to the surface directly 20160915 08:22:08< Aginor> that's a good goal 20160915 08:22:13< vultraz> and standardize them too 20160915 08:22:39< vultraz> this is why I introduced the software render context thing. 20160915 08:22:45< Aginor> but what we really need to do is to draw to fewer surfaces in general, ideally we should just draw to the screen 20160915 08:22:54< vultraz> right 20160915 08:23:07< vultraz> but how do we get there 20160915 08:23:54< vultraz> like, right now, text looks like absolute shit unless it's blitted with the custom blit function 20160915 08:23:58< Aginor> so adding a rendering api to render to a surface is doubling-down on the path that's counterproductive to getting us towards a goal where we can "simply" render our textures 20160915 08:24:16< Aginor> so let's fix the text rendering and use the sdl blit 20160915 08:24:23< vultraz> yeah, you have a point.. 20160915 08:25:26< Aginor> and if we make sure that we have the text rendering be onto a surface of the appropriate size and with the right alpha, we can then statically convert it into a texture and we're ready for hw support 20160915 08:26:46< vultraz> yeah, but we need to do that for a whole lot of stuff :/ 20160915 08:27:18< Aginor> yes 20160915 08:27:37< zookeeper> tad_, curiously, the code seems pretty weird. it looks like if the key is not preset, true is assumed. and then if it's present, nothing gets done with it so presumably it'll be false even if you specify modify_placing=true. but that's just my reading of it. 20160915 08:27:51< Aginor> so the things that are stopping us from doing hw support right now is the custom blitting implementation 20160915 08:28:07< Aginor> the undrawing pattern stops of from being even vaguely efficient about it 20160915 08:28:21< Aginor> as we are constantly copying pixels around 20160915 08:28:25< tad_> zookeeper: Just out of curiosity: what would happen is I were to add modify_placing to a scenario which was NOT randomly generated? 20160915 08:28:27 * zookeeper is nostalgically reminded of the long epic hunt for the mystery OOS of cc9cf1a8 20160915 08:28:44< vultraz> and to ALSO confuse me further is there's not just blit_surface, but also CVideo::blit_surface, which calls sdl_blitsurface.. 20160915 08:29:03< vultraz> s/is// 20160915 08:29:39< zookeeper> tad_, no idea. try it :J i mean the feature is for placing leaders on castles which have terrain favourable to their faction around them, so on most regular MP maps it would probably be pretty arbitrary. 20160915 08:29:42< Aginor> yes. that should all just be turned into the one codepath 20160915 08:29:51< vultraz> I guess we're worried about plain blit_surface 20160915 08:30:09< tad_> zookeeper: I'm looking at game_state::init because that's where [time_area] is handled and pondering adding [terrain_mask] there. 20160915 08:30:38< Aginor> sdl/utils.cpp:blit_surface is the source of all of the pain 20160915 08:30:44< vultraz> yes 20160915 08:31:04< tad_> zookeeper: And saw that option .. and could not find it in the wiki .. but you're right it's in generator as set to "true" there. 20160915 08:31:24< vultraz> but I cannot figure out what to do to surfaces to make them not need that 20160915 08:31:39< zookeeper> tad_, to support [terrain_mask] as child of [scenario]? well, okay. i take it that the only reason to do that instead of just putting it in a prestart event is to allow special locations to be overridden before side initialization takes place? 20160915 08:31:46< Aginor> as JyrkiVesterinen said, it's an alpha issue 20160915 08:32:03< vultraz> but *how does one fix that for surfaces*! 20160915 08:32:46< vultraz> modify the format masks? or what 20160915 08:32:48< tad_> zookeeper: Yes. So one could do a [unit] in a [side] and have placement= where is a map named location 20160915 08:33:47< tad_> zookeeper: I told myself I'd waste a day or so thinking about it. 20160915 08:34:23-!- atarocch [~atarocch@62.134.204.4] has joined #wesnoth-dev 20160915 08:34:37 * Aginor looks at text.cpp 20160915 08:34:45< Aginor> // The cairo surface is in CAIRO_FORMAT_ARGB32 which uses 20160915 08:34:45< Aginor> // pre-multiplied alpha. SDL doesn't use that so the pixels need to be 20160915 08:34:46< Aginor> // decoded again. 20160915 08:35:40< tad_> zookeeper: My PR on getting [terrain_mask] working is done, btw. So I could move the units to prestart and have the same effect now that [terrain_mask] actually works. 20160915 08:35:41< zookeeper> tad_, oh and if the HttT PR is ready now, i'll merge it in a few minutes... 20160915 08:36:00< loonycyborg> Aginor: can you have a look at last travis fail on campaignd_asio branch? 20160915 08:36:01< tad_> I made the changes you requested. 20160915 08:36:41< vultraz> Aginor: hmm 20160915 08:36:52< vultraz> Aginor: actually, I may have found a partial solution to the text thing 20160915 08:37:02< Aginor> the pixelformat there is smelly 20160915 08:38:15< vultraz> Aginor: if I set the text surface blend mode to SDL_BLENDMODE_NONE before using sdl_blitsurface it works 20160915 08:38:31< vultraz> only problem is it then overrides the backgrounds of butttons o_O 20160915 08:38:44< Aginor> yes, because you're not blending 20160915 08:38:48< Aginor> you do want to blend ;) 20160915 08:39:10< vultraz> but if you do blend the text looks like shit :P 20160915 08:40:58< Aginor> so fix how the surface is created 20160915 08:41:10< Aginor> there's stuff going on there that's looking rather-not-great 20160915 08:41:31< vultraz> I don't understand that code, though 20160915 08:47:18< vultraz> for the record, disabling the text "decoding" doesn't help anything 20160915 08:47:29< Aginor> nope, it doesn't 20160915 08:48:11< vultraz> but does that mean it's needed and it's broken, or it's the wrong fix for another problem 20160915 08:52:58-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .] 20160915 08:53:19< Aginor> https://www.cairographics.org/SDL/ <-- looks related 20160915 08:58:05< vultraz> t'is rather hard to read 20160915 08:58:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 08:59:31< vultraz> "You can also use cairo to composite a cairo surface with transparency to an SDL_Surface without, and thus sidestep the problem entirely." 20160915 08:59:33< vultraz> hmm 20160915 09:01:02< Aginor> that won't fly 20160915 09:01:14< Aginor> not if we want it to be forward-compatible 20160915 09:01:26< Aginor> we want a surface that's completely transparent except for the text 20160915 09:02:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds] 20160915 09:03:23< vultraz> true 20160915 09:03:33< vultraz> im trying to parse this cairosdl thing 20160915 09:03:46< vultraz> that's the thing that's hard to read 20160915 09:04:59< vultraz> maybe ill let someone else play with this 20160915 09:05:26< vultraz> Aginor: so, for now, what direction should i take with my branch. should i leave as-is? 20160915 09:05:30< Aginor> I'm wondering if that's a 1.2 thing 20160915 09:06:27< vultraz> considering it's from 2009, likely 20160915 09:07:08< Aginor> yeah 20160915 09:07:26< vultraz> so we're back to square 1 20160915 09:07:28< vultraz> :/ 20160915 09:07:42< vultraz> there seems to be precious little info about surface handling in sdl. 20160915 09:07:58< Aginor> read the source ;) 20160915 09:08:13< vultraz> D: 20160915 09:08:13< Aginor> and the headers. The headers are really good 20160915 09:09:41< Aginor> as for the branch, I still think the drawing-primitives in gui2 canvas has to go 20160915 09:10:45< vultraz> eh.... 20160915 09:11:02< vultraz> I'm conflicted 20160915 09:11:23< vultraz> it's a better implementation than what there was before 20160915 09:11:34< Aginor> taking out the sdl1 compat layer is good 20160915 09:11:41< vultraz> but you have a point about it rather being a step back 20160915 09:11:48< vultraz> in some ways.. 20160915 09:11:50< vultraz> so I'm not sure 20160915 09:12:02< vultraz> I mean, stuff like draw_solid_tinted_rectangle is just blagh 20160915 09:12:20< vultraz> and i like my new impl better.. 20160915 09:12:45< Aginor> the changes to sdl_surface goes against the idea of drawing to the screen so I don't like them either 20160915 09:13:05< Aginor> so I once again ask, why do we need to render onto an arbitrary surface? 20160915 09:13:27< Aginor> if you want those drawing primitives, why not draw them on the screen? 20160915 09:13:43< Aginor> (although rendering something that could be turned into a texture would be acceptable) 20160915 09:14:15< vultraz> because I'd have to re-implement the entire canvas system to do that 20160915 09:14:17< vultraz> wouldn't I? 20160915 09:15:43-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev 20160915 09:15:49< vultraz> id love to render the entire canvas to the back framebuffer and draw it on the next draw cycle 20160915 09:15:54< Aginor> loonycyborg: this one? https://travis-ci.org/wesnoth/wesnoth/jobs/159928805 20160915 09:16:19< vultraz> but I thought such a thing was not possible 20160915 09:16:36 * Aginor goes to look at canvas again 20160915 09:17:05< loonycyborg> Aginor: yes 20160915 09:17:42< loonycyborg> it seems that for some reason with clang it has no overload for std::chrono 20160915 09:18:00< vultraz> Aginor: all the widget impl_draw_foreground/background functions passaround a surface that's dealt with in twindow::draw 20160915 09:18:51< loonycyborg> even though exact same clang++ commands work locally for me, with my install of clang 20160915 09:19:29< Aginor> loonycyborg: I don't have clang installed... 20160915 09:20:17< loonycyborg> is same version of boost used by all travis builds? 20160915 09:20:24< zookeeper> ok, merge time... prepare for 30 commits 20160915 09:20:34< zookeeper> (no muting) 20160915 09:20:40< Aginor> loonycyborg: as far as I know 20160915 09:21:19< irker112> wesnoth: Gregory A Lundberg wesnoth:master cb074258e64d / data/campaigns/Heir_To_The_Throne/scenarios/02_Blackwater_Port.cfg: HttT S02 Haldiel rides in https://github.com/wesnoth/wesnoth/commit/cb074258e64d4c684ecdd12b7c6d4aa8db9d57d1 20160915 09:21:21< irker112> wesnoth: Gregory A Lundberg wesnoth:master 24ed9e55b3ef / data/campaigns/Heir_To_The_Throne/scenarios/02_Blackwater_Port.cfg: HttT S02 Simyr steps forward https://github.com/wesnoth/wesnoth/commit/24ed9e55b3eff5a9364ccbfde4eb31fa86bd45b7 20160915 09:21:23< irker112> wesnoth: Gregory A Lundberg wesnoth:master 1c9ef479dd64 / data/campaigns/Heir_To_The_Throne/scenarios/02_Blackwater_Port.cfg: HttT S02 Consistent team color for Sir Kaylan https://github.com/wesnoth/wesnoth/commit/1c9ef479dd64e40869d5327344158551969acda8 20160915 09:21:25< irker112> wesnoth: Gregory A Lundberg wesnoth:master 051ce00bd8ff / data/campaigns/Heir_To_The_Throne/scenarios/04_The_Bay_of_Pearls.cfg: HttT S04 Use best of Merfolk https://github.com/wesnoth/wesnoth/commit/051ce00bd8ff7bef286b7819c5566df7c9720155 20160915 09:21:27< irker112> wesnoth: Gregory A Lundberg wesnoth:master fa60cbe7f16a / data/campaigns/Heir_To_The_Throne/scenarios/04_The_Bay_of_Pearls.cfg: HttT S04 Use best of Landlubbers https://github.com/wesnoth/wesnoth/commit/fa60cbe7f16a6565cf25f9684554200ebd2d51c1 20160915 09:21:29< irker112> wesnoth: Gregory A Lundberg wesnoth:master 002fd030bd41 / data/campaigns/Heir_To_The_Throne/scenarios/04_The_Bay_of_Pearls.cfg: HttT S04 Pan and animate Bugg https://github.com/wesnoth/wesnoth/commit/002fd030bd41bc6b971367e0a04cebc84274455d 20160915 09:21:31< irker112> wesnoth: Gregory A Lundberg wesnoth:master e5bac96a6003 / data/campaigns/Heir_To_The_Throne/scenarios/05a_Muff_Malal_Peninsula.cfg: HttT S05a Use best Advisor https://github.com/wesnoth/wesnoth/commit/e5bac96a6003c65fee594f6be3a69f72dad69644 20160915 09:21:33< irker112> wesnoth: Gregory A Lundberg wesnoth:master 4e42034c7709 / data/campaigns/Heir_To_The_Throne/scenarios/08_The_Princess_of_Wesnoth.cfg: HttT S08 Adjust closing-in area https://github.com/wesnoth/wesnoth/commit/4e42034c7709b02cf79a99e99325347220a3caa1 20160915 09:21:35< irker112> wesnoth: Gregory A Lundberg wesnoth:master 6cef43770b20 / data/campaigns/Heir_To_The_Throne/scenarios/08_The_Princess_of_Wesnoth.cfg: HttT S08 Pan to duelist https://github.com/wesnoth/wesnoth/commit/6cef43770b20ade1dd851e23cf2c9f42970ed3aa 20160915 09:21:37< irker112> wesnoth: Gregory A Lundberg wesnoth:master e0283a6addf0 / data/campaigns/Heir_To_The_Throne/scenarios/09_The_Valley_of_Death.cfg: HttT S09 Victory before Moremirmu arrives https://github.com/wesnoth/wesnoth/commit/e0283a6addf016737b5ec4ec0ff4dc3abc038609 20160915 09:21:39< irker112> wesnoth: Gregory A Lundberg wesnoth:master 9c3301a78bd0 / data/campaigns/Heir_To_The_Throne/scenarios/10_Gryphon_Mountain.cfg: HttT S10 Change objectives https://github.com/wesnoth/wesnoth/commit/9c3301a78bd014a0fed2d7d936ec58681fbb2257 20160915 09:21:41< irker112> wesnoth: Gregory A Lundberg wesnoth:master c952517875e1 / data/campaigns/Heir_To_The_Throne/scenarios/11_The_Ford_of_Abez.cfg: HttT S11 Improved reinforcements https://github.com/wesnoth/wesnoth/commit/c952517875e184f5dfd3eade499c2d89ddbbcd97 20160915 09:21:43< irker112> wesnoth: Gregory A Lundberg wesnoth:master 117259e35b23 / data/campaigns/Heir_To_The_Throne/scenarios/11_The_Ford_of_Abez.cfg: HttT S11 Hide side https://github.com/wesnoth/wesnoth/commit/117259e35b2344c39ec896e42cb7c8658ef632a7 20160915 09:21:45< irker112> wesnoth: Gregory A Lundberg wesnoth:master 8ae8cb0384f2 / data/campaigns/Heir_To_The_Throne/scenarios/13_The_Dwarven_Doors.cfg: HttT S13 Use best outlaw https://github.com/wesnoth/wesnoth/commit/8ae8cb0384f2f010c4d6e15840bae64948af0712 20160915 09:21:47< irker112> wesnoth: Gregory A Lundberg wesnoth:master 326e36e16bc9 / data/campaigns/Heir_To_The_Throne/scenarios/13_The_Dwarven_Doors.cfg: HttT S13 Increased comment chance https://github.com/wesnoth/wesnoth/commit/326e36e16bc995b49673f91c051430f6566204c0 20160915 09:21:49< irker112> wesnoth: Gregory A Lundberg wesnoth:master 92290d3bbc50 / data/campaigns/Heir_To_The_Throne/scenarios/13_The_Dwarven_Doors.cfg: HttT S13 Use best whiner https://github.com/wesnoth/wesnoth/commit/92290d3bbc5084bdbaf0cd9001a3a5f26772e7bf 20160915 09:21:51< irker112> wesnoth: Gregory A Lundberg wesnoth:master 853144a39aee / data/campaigns/Heir_To_The_Throne/scenarios/13_The_Dwarven_Doors.cfg: HttT S13 Hide side https://github.com/wesnoth/wesnoth/commit/853144a39aee9b4756b695f8eb2512e160b16ea2 20160915 09:21:53< irker112> wesnoth: Gregory A Lundberg wesnoth:master 69c2bcd49995 / data/campaigns/Heir_To_The_Throne/scenarios/14_Plunging_Into_the_Darkness.cfg: HttT S14 Consistent objectives https://github.com/wesnoth/wesnoth/commit/69c2bcd49995e5d46c7a0f2d0902d2025b201d57 20160915 09:21:55< irker112> wesnoth: Gregory A Lundberg wesnoth:master d9bd9f9bb2c3 / data/campaigns/Heir_To_The_Throne/scenarios/14_Plunging_Into_the_Darkness.cfg: HttT S14 Hide side https://github.com/wesnoth/wesnoth/commit/d9bd9f9bb2c372ca06022e9709000af2b74d691e 20160915 09:21:57< irker112> wesnoth: Gregory A Lundberg wesnoth:master ac3db7f036e1 / data/campaigns/Heir_To_The_Throne/scenarios/17_Scepter_of_Fire.cfg: HttT S17 No need to change objectives https://github.com/wesnoth/wesnoth/commit/ac3db7f036e19a44c12ba9a00d25d7e05cf6dccd 20160915 09:21:59< irker112> wesnoth: Gregory A Lundberg wesnoth:master c8ca26c833d8 / data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: HttT S19a Improve Flaming Sword https://github.com/wesnoth/wesnoth/commit/c8ca26c833d8b71fcd6818c0dc5da973ee4bbacd 20160915 09:22:01< irker112> wesnoth: Gregory A Lundberg wesnoth:master ef641d6ed3df / data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: HttT S19b Improve Void Armor https://github.com/wesnoth/wesnoth/commit/ef641d6ed3df25818006765a346dea22ea17c2b9 20160915 09:22:03< irker112> wesnoth: Gregory A Lundberg wesnoth:master b452760fb5b3 / data/campaigns/Heir_To_The_Throne/scenarios/ (18_A_Choice_Must_Be_Made.cfg 19c_Cliffs_of_Thoria.cfg): HttT S19c Follow the leader https://github.com/wesnoth/wesnoth/commit/b452760fb5b3a4b91e0b5ee56ac9626b02bb65fe 20160915 09:22:05< irker112> wesnoth: Gregory A Lundberg wesnoth:master f836609ba5ca / data/campaigns/Heir_To_The_Throne/scenarios/19c_Cliffs_of_Thoria.cfg: HttT S19c Less delay https://github.com/wesnoth/wesnoth/commit/f836609ba5cacc15968e75a55f0b046247eead70 20160915 09:22:07< irker112> wesnoth: Gregory A Lundberg wesnoth:master e9feb4c5780b / data/campaigns/Heir_To_The_Throne/scenarios/19c_Cliffs_of_Thoria.cfg: HttT S19c Hide sides https://github.com/wesnoth/wesnoth/commit/e9feb4c5780b27fba366e9404e8627adec531ee6 20160915 09:22:09< irker112> wesnoth: Gregory A Lundberg wesnoth:master 9b80ea96890b / data/campaigns/Heir_To_The_Throne/ (4 files in 2 dirs): HttT bigmap Fix bug: Wrong defaults https://github.com/wesnoth/wesnoth/commit/9b80ea96890bc34f1c5ecaa44ce174d92a61922d 20160915 09:22:11< irker112> wesnoth: Gregory A Lundberg wesnoth:master ecf4a1366ac7 / data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: HttT S19a Fix bug: Paladin can use sword https://github.com/wesnoth/wesnoth/commit/ecf4a1366ac7ff507fe203b253d813ea4ed09568 20160915 09:22:13< irker112> wesnoth: Gregory A Lundberg wesnoth:master d8dad92bd906 / data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: HttT S19b Use unit for Void Armor https://github.com/wesnoth/wesnoth/commit/d8dad92bd906e7eff9407ac6a598c76195e202a3 20160915 09:22:15< irker112> wesnoth: Gregory A Lundberg wesnoth:master 533e87257885 / data/campaigns/Heir_To_The_Throne/utils/httt_utils.cfg: HttT S19a Fix bug: Not translatable https://github.com/wesnoth/wesnoth/commit/533e872578853d216b2ededf6083823f9e1dbe30 20160915 09:22:17< irker112> wesnoth: Gregory A Lundberg wesnoth:master 7778e6f45c0e / data/campaigns/Heir_To_The_Throne/ (29 files in 2 dirs): HttT Debug choose_level works https://github.com/wesnoth/wesnoth/commit/7778e6f45c0e0bb7d19075b9a0ab15463b183a74 20160915 09:22:19< irker112> wesnoth: Lari Nieminen wesnoth:master c11498ab8902 / data/campaigns/Heir_To_The_Throne/ (30 files in 2 dirs): Merge pull request #730 from GregoryLundberg/GL_HttT_general_improvements https://github.com/wesnoth/wesnoth/commit/c11498ab89029201c52c8d5318c69c3e8345448c 20160915 09:23:49< zookeeper> tad_, i can start pushing in the other HttT PR's whenever you've rebased them 20160915 09:24:17< tad_> zookeeper: working on it 20160915 09:26:05< Aginor> vultraz: ok, keep them then 20160915 09:27:18< Aginor> or actually, don't 20160915 09:28:10< Aginor> http://www.ferzkopp.net/wordpress/2016/01/02/sdl_gfx-sdl2_gfx/ 20160915 09:28:37< Aginor> https://sourceforge.net/projects/sdl2gfx/ even 20160915 09:29:33< vultraz> I'll take a look 20160915 09:30:50< Aginor> those primitives are better, and faster 20160915 09:31:05< Aginor> better edge case handling 20160915 09:32:00< vultraz> so you recommend using that? 20160915 09:32:11< vultraz> how would that factor into our larger vision of hw accl? 20160915 09:32:18< tad_> zookeeper: I believe all remaining HttT PR are properly rebased and ready. Each should be ONE commit with a name fitting the PR. 20160915 09:32:47< Aginor> vultraz: that kind of canvas is never going to be static 20160915 09:33:01< vultraz> i don't follow 20160915 09:33:12< Aginor> so make it once as a surface, convert to a texture, render the texture 20160915 09:33:37< Aginor> when the user adds things, re-do the surface and update the texture 20160915 09:33:41< vultraz> so you suggest rendering the entire canvas as a texture 20160915 09:33:53< Aginor> (mark it as a streaming texture for better performance) 20160915 09:33:56< Aginor> yes 20160915 09:34:01< vultraz> and passing it to the window renderer 20160915 09:34:33< vultraz> this will be work 20160915 09:34:42< Aginor> I don't particurly like it, but unless you keep track of the drawing operations you want to perform, that's how it has to be 20160915 09:35:02< vultraz> can I at least merge the first 4 commits from the branch, since those are less related to this? 20160915 09:35:30 * Aginor goes to review those commits 20160915 09:36:22< irker112> wesnoth: loonycyborg wesnoth:campaignd_asio 7f204f026a0e / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): Explicitly instantiate steady clock with std::chrono https://github.com/wesnoth/wesnoth/commit/7f204f026a0ee44065446575676bf09269e4c2dd 20160915 09:37:36< Aginor> 633d7d1c039436959ce04544c8fa263ce5f57ae4 to ed8c8fbd396eb0a9643a6993ea9d550033b35adb looks worth doing 20160915 09:37:53< Aginor> I'm less sure about ed8c8fbd396eb0a9643a6993ea9d550033b35adb 20160915 09:40:31< vultraz> er, you mentioned ed8c8fbd3 twice 20160915 09:41:04< Aginor> because copy paste using the github copy-button doesn't work 20160915 09:41:20< Aginor> was supposed to be 646ee1dc68a6b9f439f424160db1832143441a8f that I'm less sure about 20160915 09:41:26-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 09:41:27< travis-ci> wesnoth/wesnoth#10939 (master - c11498a : Lari Nieminen): The build is still failing. 20160915 09:41:28< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160112504 20160915 09:41:28-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 09:42:26< vultraz> ah 20160915 09:42:38< vultraz> that I did because I want to remove scoped_resource 20160915 09:42:45< Aginor> yes 20160915 09:42:49< vultraz> surface is the last bastion 20160915 09:43:01< Aginor> I wonder if we wouldn't be able to make it be a shared_ptr instead 20160915 09:43:29< Aginor> although that might break things horribly since we are passing around the one reference to the screen that is blitted onto 20160915 09:43:33< vultraz> i did try that but uh... cannot have custom refcounting 20160915 09:43:46< Aginor> why do we need the custom refcounting though? 20160915 09:44:02< irker112> wesnoth: Gregory A Lundberg wesnoth:master 5fcc2c232f40 / data/campaigns/Heir_To_The_Throne/scenarios/05b_Isle_of_the_Damned.cfg: HttT S05b Randomize temples https://github.com/wesnoth/wesnoth/commit/5fcc2c232f4090c206a872b257cc68da14ba9408 20160915 09:44:04< irker112> wesnoth: Lari Nieminen wesnoth:master d16ce1b300a4 / data/campaigns/Heir_To_The_Throne/scenarios/05b_Isle_of_the_Damned.cfg: Merge pull request #732 from GregoryLundberg/GL_HttT_S05b_randomize_temples https://github.com/wesnoth/wesnoth/commit/d16ce1b300a441b4b5a13171c5c205d7f2590f23 20160915 09:45:12< vultraz> pretty sure it's so multiple surface objects can refer to the same SDL_Surface and it's only destroyed by FreeSurface if no-one owns it. 20160915 09:46:09< vultraz> i tried using a shared_ptr and it crashed wesnoth immediately 20160915 09:46:19< vultraz> quite possibly i did it wrong, though :P 20160915 09:46:31< Aginor> that's probably because of what I stated above ;) 20160915 09:48:30< vultraz> so what I propose is to merge those 4 commits, likely roll back the other 2 and start working on something using sdl2gfx 20160915 09:48:44< Aginor> that makes sense 20160915 09:49:06< Aginor> where I think we should try to get rid of number 4 20160915 09:49:27< vultraz> hm? 20160915 09:49:41< vultraz> what do you mean 20160915 09:49:42< Aginor> we *should* be able to use a shared pointer ;) 20160915 09:49:57< vultraz> true 20160915 09:50:02< Aginor> but it's probably the handling for the screen surface that's causing grief 20160915 09:50:17< vultraz> but i hope you don't object to the self-rolled functionality for now while we purge scoped_resource 20160915 09:50:48< Aginor> run it through valgrind before merging to make sure you're not leaking memory 20160915 09:51:13< vultraz> oh deer. 20160915 09:51:40< vultraz> valgring I have not, nor have used D: 20160915 09:52:27< Aginor> how do you memory leak detection? 20160915 09:53:13< vultraz> I don't 20160915 09:54:15< Aginor> it's a good thing to learn how to do 20160915 09:54:19< tad_> run wesnoth through valgrind? That should be VERY interesting 20160915 09:54:30< Aginor> tad_: it's actually pretty good 20160915 09:54:35< Aginor> tad_: I do that every now and then 20160915 09:54:44< Aginor> tad_: callgrind too 20160915 09:55:32< tad_> Aginor: So the memory growth I see as I play isn't leaks? 20160915 09:56:28< Aginor> tad_: sadly, not as far as I can tell 20160915 09:56:49< Aginor> tad_: it's just keeping everything in memory that it ever touches 20160915 09:57:31< tad_> Seems odd, then. I play and it grows and grows and I stop, and load from a save and it's back down low. So it never releases the memory for past scenario? 20160915 09:58:38< Aginor> no, it tends to cache everything 20160915 09:59:13< tad_> oy. we need to find that someday and have it free stuff up. 20160915 09:59:58< Aginor> we need to speed up the wml parser too, to make it feel less like a need to cache everything 20160915 10:01:07< vultraz> making it not a string preprocessor would help 20160915 10:02:15< Aginor> hmm 20160915 10:02:29< Aginor> why are we using cairo as our text rendering library? 20160915 10:04:32< zookeeper> we need a file listing all these special characters we use. like the ellipsis, proper-kind-of-single-quote, etc. it's always a huge pain to try to find a copypaste source. 20160915 10:04:39< zookeeper> (in text, that is) 20160915 10:05:05< tad_> I'll say. 20160915 10:06:07< zookeeper> hmh, i guess we don't actually use the ellipsis character after all, only "..."? well, that helps 20160915 10:08:32-!- Kwandulin [~Miranda@p200300760F2C71D3C1B8DFCCD1DCB168.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 20160915 10:13:03-!- iceiceice [~chris@unaffiliated/iceiceice] has quit [Ping timeout: 240 seconds] 20160915 10:21:24< tad_> zookeeper: A quick check and we do have a dozen or so characters in data/campaigns/*.cfg which are outside the ANSI/ASCII range (0x01 .. 0x7E). 20160915 10:21:54< zookeeper> pastebinnable? 20160915 10:22:10< tad_> Interestingly, we do use \t .. probably need to find and fix that 20160915 10:22:55< tad_> Not yet. I did it with od so I have the byte codes .. need to do a bit more to save them as UTF-8 20160915 10:23:30< tad_> They're octal. And I could puzzle the UTF-8 out by hand but there's probably a better way 20160915 10:24:23< zookeeper> \t doesn't seem to be used for indentation anywhere, just as preceding whitespace for comments or in lua code. 20160915 10:25:06< tad_> probably. only appears 233 times in all of data/campaigns/*.cfg 20160915 10:26:24< Aginor> vultraz: decode_pixel is indeed needed, it un-premultiplies the alpha values of the bitmap 20160915 10:33:29< vultraz> ok ,good to know 20160915 10:33:52< Aginor> anyway, it's late and I'm tired 20160915 10:33:57< vultraz> Aginor: so do I absolutely need to run valgrind on the pr? 20160915 10:33:57< Aginor> so I shall disappear 20160915 10:34:38< Aginor> vultraz: you're messing with memory management of one of the more frequently churned objects, I would suggest making sure you don't introduce leaks 20160915 10:35:18< Aginor> checking and fixing before it's merged and affects everyone/derails the release is the prudent thing to do 20160915 10:37:37< Aginor> vultraz: good job 20160915 10:38:00< vultraz> er, did I break something? 20160915 10:38:25< Aginor> vultraz: no, you've done good with tidying up a bunch of cruft 20160915 10:39:11< Aginor> (it's praise, not sarcasm ;) ) 20160915 10:39:22< vultraz> oh, thanks :D 20160915 10:40:26< tad_> zookeeper: The wife just realized she pulled an all-nighter sewing and so I'm off to bed. I'll check the PRs for more comments and address your issues when we get up later today ... night 20160915 10:40:40< zookeeper> sure 20160915 10:40:45-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed] 20160915 10:44:16-!- Kwandulin [~Miranda@p200300760F2C71D3F0415E6E27F98A27.dip0.t-ipconnect.de] has joined #wesnoth-dev 20160915 10:47:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 10:51:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 265 seconds] 20160915 10:59:09< Aginor> good night guys 20160915 11:15:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20160915 11:24:48-!- RatArmy [~RatArmy@240f:b3:88e3:1:224:a5ff:fe23:83eb] has joined #wesnoth-dev 20160915 11:35:36-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 11:35:37< travis-ci> wesnoth/wesnoth#10947 (campaignd_asio - 7f204f0 : loonycyborg): The build passed. 20160915 11:35:37< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160115520 20160915 11:35:37-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 11:35:49< loonycyborg> :O 20160915 11:41:13< vultraz> :O 20160915 11:42:05< matthiaskrgr> :O 20160915 11:54:44-!- louis94 [~~louis94@91.178.241.97] has joined #wesnoth-dev 20160915 12:02:32< loonycyborg> vultraz: that's weird, I can't repro the travis failure on master 20160915 12:02:53< vultraz> the thing about swap? 20160915 12:05:32-!- RatArmy [~RatArmy@240f:b3:88e3:1:224:a5ff:fe23:83eb] has quit [Quit: Leaving] 20160915 12:06:10< loonycyborg> yes 20160915 12:06:17-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 12:06:18< travis-ci> wesnoth/wesnoth#10949 (master - d16ce1b : Lari Nieminen): The build is still failing. 20160915 12:06:18< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160117134 20160915 12:06:18-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 12:08:21< vultraz> yea, no one can seem to repro it 20160915 12:08:28< loonycyborg> vultraz: I had an idea, maybe try to replace it with std::swap 20160915 12:08:43< vultraz> it's likely it's unimplemented in gcc 4.8 or something 20160915 12:08:52< vultraz> since it's a c++11 feature. 20160915 12:08:55< vultraz> I could try that 20160915 12:11:35< irker112> wesnoth: Charles Dang wesnoth:master a4a1ae49f146 / src/gui/dialogs/end_credits.cpp: Another attempt to satisfy travis https://github.com/wesnoth/wesnoth/commit/a4a1ae49f146edddc499551223abcb6bfb1810a2 20160915 12:24:18-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20160915 12:25:13-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has joined #wesnoth-dev 20160915 12:28:45-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 12:28:46< travis-ci> gfgtdf/wesnoth-old#684 (location_refeactor - bd81731 : gfgtdf): The build failed. 20160915 12:28:46< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/160148588 20160915 12:28:46-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 12:31:16< celticminstrel> ...location refactor? 20160915 12:31:48-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 12:31:49< travis-ci> wesnoth/wesnoth#10950 (master - a4a1ae4 : Charles Dang): The build is still failing. 20160915 12:31:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160146396 20160915 12:31:49-!- travis-ci [~travis-ci@ec2-54-162-120-232.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 12:32:45< vultraz> :( 20160915 12:33:50< vultraz> le sigh 20160915 12:33:55< vultraz> it doesn't like std::swap 20160915 12:34:10< celticminstrel> Oh, you tried switching from .swap to std::swap? 20160915 12:34:11< vultraz> travis is an idiot 20160915 12:34:13< vultraz> yes 20160915 12:34:28< celticminstrel> I wouldn't've expected that to make any difference, for reasons I elucidated earlier. 20160915 12:34:35< vultraz> so what do we do 20160915 12:34:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 12:34:40< celticminstrel> (If there's no specialization, std::swap relies on copying.) 20160915 12:34:48< celticminstrel> (And streams are non-copyable.) 20160915 12:34:56< vultraz> obviously travis doesn't have the specialization of swap for stringsteams in 4.8 20160915 12:35:10< celticminstrel> Obviously. 20160915 12:35:47< celticminstrel> I think what we do should be simply finding a different way. 20160915 12:36:29< vultraz> or we can bump travis :) 20160915 12:36:53< celticminstrel> I seem to recall that that's not an option. 20160915 12:37:33< vultraz> if stringsteams don't copy then this is likely the most efficient way to do it 20160915 12:37:57< celticminstrel> "it" 20160915 12:38:06< celticminstrel> Please explain what "it" is. 20160915 12:38:23< vultraz> moving the highlighted credits section to the beginning 20160915 12:39:09< celticminstrel> Speaking of which, do you see credits for all installed campaigns even when just viewing them from the titlescreen? 20160915 12:39:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds] 20160915 12:39:56< vultraz> yes 20160915 12:41:15< vultraz> what compilers would we lose if we moved travis to gcc 5? 20160915 12:42:03< celticminstrel> Oh huh, it's true even in 1.12. 20160915 12:42:47< celticminstrel> For some reason I have shadow m's campaigns installed even though I've never played UtBS. 20160915 12:43:13< celticminstrel> BTW, if I recall the new credits correctly, you need an extra line break between sections. 20160915 12:43:15< vultraz> i assume you can build master fine? 20160915 12:44:06< celticminstrel> I haven't checked yet. I started refactoring menus and can't build because of that. 20160915 12:44:14< vultraz> I see 20160915 12:44:26< celticminstrel> I guess the old credits loop, so that's something different about the new ones... 20160915 12:45:14< celticminstrel> (In other words, if you already added an extra line break, then ignore what I just said.) 20160915 12:45:24< vultraz> no 20160915 12:45:28< vultraz> i cannot add the linebreak 20160915 12:45:36< vultraz> without making the credits not appear 20160915 12:45:38< celticminstrel> It needs to be done. 20160915 12:47:48< irker112> wesnoth: Charles Dang wesnoth:master e2614186910c / src/gui/dialogs/end_credits.cpp: Revert "Another attempt to satisfy travis" https://github.com/wesnoth/wesnoth/commit/e2614186910c756412be44d7f55e1c929eea9cba 20160915 12:48:13< celticminstrel> I assume installing any addon campaig will also make the credits not appear. 20160915 12:48:24< celticminstrel> ^+n 20160915 12:48:35< vultraz> i have multiple addon campaigns installed 20160915 12:48:45< celticminstrel> Ahah, so that's the problem. 20160915 12:48:59< vultraz> ...what 20160915 12:49:01< celticminstrel> That implies it would work for me if the line break was added. 20160915 12:49:10< celticminstrel> Mind you, it still needs to work if addon campaigns are installed. 20160915 12:49:21< celticminstrel> Just how many do you have installed on 1.13? 20160915 12:49:34< vultraz> at least 5 or so 20160915 12:49:40< celticminstrel> I see... 20160915 12:51:11< vultraz> and yes, it works without addons 20160915 12:55:55< celticminstrel> I wonder if we can disable window scrollbars in MP Create... it could lead to "doesn't fit on screen" crashes if not careful though... 20160915 12:56:32< celticminstrel> Apparently the developer of Braid recommends sticking with MSVC 2013 instead of 2015. 20160915 12:57:24< vultraz> why? 20160915 12:57:33< vultraz> how is that a good idea in *any* way 20160915 12:57:34< celticminstrel> Didn't say. 20160915 12:57:41< vultraz> 2013 has less c++11 support 20160915 12:57:44< vultraz> and is older 20160915 12:57:58< celticminstrel> True, less C++11, but I guess he finds it worse in other ways. 20160915 12:58:23< celticminstrel> (I wonder if you could take the cl.exe from MSVC 2015 and hook it into MSVC 2013.) 20160915 12:58:57< celticminstrel> (For my own stuff though, I'll probably be using clang in MSVC 2013, so the C++11 issue doesn't exist.) 20160915 13:01:01-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has quit [Quit: leaving] 20160915 13:02:21< irker112> wesnoth: Charles Dang wesnoth:master 61f1c83073e1 / data/gui/window/end_credits.cfg src/gui/dialogs/end_credits.cpp: End Credits: few formatting changes https://github.com/wesnoth/wesnoth/commit/61f1c83073e13bfb395fd1f8b752c0f4527ce8f2 20160915 13:02:25< vultraz> there's your newlines 20160915 13:02:57< vultraz> unfortunately i have now noticed the text is perfectly centered relative to the button and it's gonna nag me ._. 20160915 13:03:26< vultraz> also made it so [about] tags with no [entry] children don't show anything 20160915 13:03:40-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20160915 13:03:41< vultraz> can't think of any case someone would want an empty heading showing 20160915 13:04:03< vultraz> (in fact I think I saw one such heading roll by in the mainline section) 20160915 13:04:54< vultraz> well this is interesting 20160915 13:05:22< vultraz> wesnoth will crash on exit if the text doesn't show and you use the scrollwheel 20160915 13:05:25< celticminstrel> I dunno, someone might want an empty heading. 20160915 13:05:46< celticminstrel> Friulian Translation is probably the empty heading you saw roll by. 20160915 13:06:02< vultraz> something like that, yeah 20160915 13:06:32< vultraz> god dammit who made scrollbar placement throw an exception it makes debugging impossible 20160915 13:06:40-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 13:06:41< travis-ci> wesnoth/wesnoth#10952 (master - e261418 : Charles Dang): The build is still failing. 20160915 13:06:41< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160154803 20160915 13:06:41-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 13:07:15< vultraz> annnddd now it went away 20160915 13:07:17< vultraz> hmm 20160915 13:07:47 * vultraz shrugs 20160915 13:07:48< vultraz> oh well 20160915 13:07:54< vultraz> non-determinate 20160915 13:18:24-!- louis94 [~~louis94@91.178.241.97] has quit [Ping timeout: 276 seconds] 20160915 13:19:04< Rhonda> How about this mockup for our website: http://cat.wesnoth.org.meowbify.com/ 20160915 13:22:47-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 13:22:48< travis-ci> wesnoth/wesnoth#10953 (master - 61f1c83 : Charles Dang): The build is still failing. 20160915 13:22:48< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160158255 20160915 13:22:48-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 13:28:16-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev 20160915 13:32:51-!- louis94 [~~louis94@91.178.241.97] has joined #wesnoth-dev 20160915 13:36:15< irker112> wesnoth: loonycyborg wesnoth:master 1aac722378ec / src/ (7 files in 2 dirs): Split off server_base class from server class in order to reuse it for campaignd https://github.com/wesnoth/wesnoth/commit/1aac722378ec1f3abf3cd18451a791b2b56eac69 20160915 13:36:17< irker112> wesnoth: loonycyborg wesnoth:master 358bf3f9f8d2 / / (11 files in 4 dirs): Switch campaignd to use asio by reusing wesnothd code via server_base class https://github.com/wesnoth/wesnoth/commit/358bf3f9f8d2ed17403961c4a1cac2d499ebf133 20160915 13:36:19< irker112> wesnoth: loonycyborg wesnoth:master bd304cb307b3 / src/ (5 files in 2 dirs): Factor server start into separate function https://github.com/wesnoth/wesnoth/commit/bd304cb307b32fe198c39f7e7370d6c695fd5142 20160915 13:36:21< irker112> wesnoth: loonycyborg wesnoth:master 31bbedd94e1b / / (38 files in 22 dirs): Merge branch 'master' into campaignd_asio https://github.com/wesnoth/wesnoth/commit/31bbedd94e1b3529499953817022dda8785285b2 20160915 13:36:23< irker112> wesnoth: loonycyborg wesnoth:master be47acce267b / src/campaign_server/campaign_server.cpp: Fix ostringstream misuse https://github.com/wesnoth/wesnoth/commit/be47acce267b588bcd9444c432fe1afa094c7799 20160915 13:36:25< irker112> wesnoth: loonycyborg wesnoth:master 702a98d50920 / src/campaign_server/campaign_server.cpp: Add missing newlines to log messages https://github.com/wesnoth/wesnoth/commit/702a98d50920d61d19ccea67b993d45ea9072dbf 20160915 13:36:27< irker112> wesnoth: loonycyborg wesnoth:master 977e76d7b137 / src/campaign_server/campaign_server.cpp: Made send_message and send_error properly maintain connection https://github.com/wesnoth/wesnoth/commit/977e76d7b137fd056a866f3678cbbd5b0f259566 20160915 13:36:29< irker112> wesnoth: loonycyborg wesnoth:master 7a09f0fd946d / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): Reimplement flushing config to disk every 10 minutes https://github.com/wesnoth/wesnoth/commit/7a09f0fd946d4be03514750cdd2bcd4704054012 20160915 13:36:31< irker112> wesnoth: loonycyborg wesnoth:master 0039bb544251 / src/campaign_server/campaign_server.cpp: Cleanup https://github.com/wesnoth/wesnoth/commit/0039bb544251b1eb8cc55806a6f1065118c6aed4 20160915 13:36:33< irker112> wesnoth: loonycyborg wesnoth:master e33f62237788 / src/team.hpp: Fix breakage caused by dropping sdl dep on some files https://github.com/wesnoth/wesnoth/commit/e33f622377889e9fc2b9aa54620e22af4b674db2 20160915 13:36:35< irker112> wesnoth: loonycyborg wesnoth:master 8710ec2bbc6f / src/server/server_base.cpp: Handle asio::error::eof to properly log client initiated disconnects https://github.com/wesnoth/wesnoth/commit/8710ec2bbc6f9a2dd674c5f024bcd68a402b5a9d 20160915 13:36:37< irker112> wesnoth: loonycyborg wesnoth:master dafc0dff6205 / src/campaign_server/campaign_server.cpp: Ensure that client disconnects get logged https://github.com/wesnoth/wesnoth/commit/dafc0dff6205c9fe94be0c5e55a6288c7042af5b 20160915 13:36:39< irker112> wesnoth: loonycyborg wesnoth:master e6bc22504025 / src/campaign_server/campaign_server.hpp: Fix accidental override of port_ member https://github.com/wesnoth/wesnoth/commit/e6bc225040256dcab14ce1b4521a0cf12f78cac6 20160915 13:36:41< irker112> wesnoth: loonycyborg wesnoth:master 2955a24b4481 / src/campaign_server/campaign_server.cpp: Properly shut down in response to fifo command https://github.com/wesnoth/wesnoth/commit/2955a24b44813a9ea6322f8fb53b710186aa4ba7 20160915 13:36:43< irker112> wesnoth: loonycyborg wesnoth:master 03f1a62a978f / src/ (campaign_server/campaign_server.cpp server/server_base.hpp): Properly recreate fifo only when necessary when reloading config https://github.com/wesnoth/wesnoth/commit/03f1a62a978f1820fd0e4be5294f21c7323dda05 20160915 13:36:45< irker112> wesnoth: loonycyborg wesnoth:master 4c616e010af3 / src/ (6 files in 2 dirs): Reindent code to be consistent with wesnoth coding standards https://github.com/wesnoth/wesnoth/commit/4c616e010af3c02c0fc8d2bf2135e7a9e7a0ffe2 20160915 13:36:47< irker112> wesnoth: loonycyborg wesnoth:master 1c90221527fe / / (3038 files in 234 dirs): Merge branch 'master' into campaignd_asio https://github.com/wesnoth/wesnoth/commit/1c90221527fe0549c843b2e8e023d2cb758c980e 20160915 13:36:49< irker112> wesnoth: Celtic Minstrel wesnoth:master 404c6ca75414 / / (2 files in 2 dirs): Fix XCode wesnothd build https://github.com/wesnoth/wesnoth/commit/404c6ca754140ca4a8979e0a204d47dad38151f3 20160915 13:36:51< irker112> wesnoth: Celtic Minstrel wesnoth:master d0b8b471c5c7 / / (2 files in 2 dirs): Add campaignd target to XCode project https://github.com/wesnoth/wesnoth/commit/d0b8b471c5c754c8a972aa712d3ad0bd5469387e 20160915 13:36:53< irker112> wesnoth: loonycyborg wesnoth:master 47a85151c649 / src/CMakeLists.txt: Update cmake with asio campaignd port sources https://github.com/wesnoth/wesnoth/commit/47a85151c649a56fcad525b543a2cbbe0bfbd474 20160915 13:36:55< irker112> wesnoth: loonycyborg wesnoth:master b1d110b95eb9 / src/ (network.cpp network.hpp network_worker.cpp network_worker.hpp tests/main.cpp): Remove old networking stack https://github.com/wesnoth/wesnoth/commit/b1d110b95eb941e921c82ddd9577a1d5e4048bc0 20160915 13:36:57< irker112> wesnoth: loonycyborg wesnoth:master d81e426ac5e4 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/d81e426ac5e40d52ff57a9ad80ccc55a8b3bab25 20160915 13:36:59< irker112> wesnoth: loonycyborg wesnoth:master 015c1ee3becd / src/campaign_server/campaign_server.cpp: Added explicit include for std::chrono https://github.com/wesnoth/wesnoth/commit/015c1ee3becd910dfd8072c3ce3348db4936bfb8 20160915 13:37:01< irker112> wesnoth: loonycyborg wesnoth:master b52571d8d9f6 / src/tests/main.cpp: Remove use of network::error missed due to conditional compilation https://github.com/wesnoth/wesnoth/commit/b52571d8d9f60fe9ede9445e80f8dac2ce6ae356 20160915 13:37:03< irker112> wesnoth: loonycyborg wesnoth:master 7f204f026a0e / src/campaign_server/ (campaign_server.cpp campaign_server.hpp): Explicitly instantiate steady clock with std::chrono https://github.com/wesnoth/wesnoth/commit/7f204f026a0ee44065446575676bf09269e4c2dd 20160915 13:37:05< irker112> wesnoth: loonycyborg wesnoth:master c16adcd8c9e4 / / (22 files in 6 dirs): Merge branch 'campaignd_asio' https://github.com/wesnoth/wesnoth/commit/c16adcd8c9e4959a6ac7e39d7801178a490598b1 20160915 13:37:17< loonycyborg> shadowm: it's merged 20160915 13:37:37-!- gfgtdf [~chatzilla@x4e368915.dyn.telefonica.de] has joined #wesnoth-dev 20160915 13:37:44< celticminstrel> Wait, did you rebase it into master? 20160915 13:37:55< celticminstrel> Oh, no, I see the merge commit, okay. 20160915 13:38:17< celticminstrel> I don't suppose you did anything to account for systems without sendfile? 20160915 13:39:38< loonycyborg> nope, not yet 20160915 13:40:06< loonycyborg> in practice all our campaignds run on linux servers atm 20160915 13:40:45< irker112> wesnoth: Bär Halberkamp wesnoth:master 6cc92e26245f / data/campaigns/Eastern_Invasion/scenarios/01_The_Outpost.cfg: EI: Make S1 slightly harder after Dacyn reappears https://github.com/wesnoth/wesnoth/commit/6cc92e26245fcbae21efef77468955aed5f62f34 20160915 13:40:55< celticminstrel> Yeah, not high priority. 20160915 13:41:02< gfgtdf> on windows there is the TransmitFile function which might do the same. 20160915 13:42:44< loonycyborg> and I have no idea why I would prefer a rebase to a merge 20160915 13:43:03< celticminstrel> Merge is probably preferable in this case. 20160915 13:43:19< loonycyborg> it seems it would result in requiring me to repeat all the work I did to resolve conflicts 20160915 13:45:27-!- louis94 [~~louis94@91.178.241.97] has quit [Ping timeout: 265 seconds] 20160915 13:56:20-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 13:56:21< travis-ci> wesnoth/wesnoth#10954 (master - c16adcd : loonycyborg): The build is still failing. 20160915 13:56:21< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160166527 20160915 13:56:21-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 13:59:12-!- Kwandulin [~Miranda@p200300760F2C71D3F0415E6E27F98A27.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 20160915 14:05:19-!- Bonobo [~Bonobo@2001:44b8:254:3200:e51f:7dcf:bc76:9e6e] has quit [Quit: Leaving] 20160915 14:08:54-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 14:08:55< travis-ci> wesnoth/wesnoth#10955 (master - 6cc92e2 : Bär Halberkamp): The build is still failing. 20160915 14:08:55< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160167905 20160915 14:08:55-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 14:08:58< celticminstrel> Support for non-Linux systems in campaignd would make it possible for non-Linux devs to work on it though, so I think it's worthwhile. 20160915 14:14:18-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20160915 14:24:03< loonycyborg> celticminstrel: would be cool if people contributed their own system-specific implementations 20160915 14:24:21< celticminstrel> I probably will... eventually. 20160915 14:28:24< vultraz> so SDL_Net is officially gone now? 20160915 14:28:31< celticminstrel> Think so. 20160915 14:28:36< vultraz> :D 20160915 14:28:53< celticminstrel> Someone please update INSTALL if it hasn't been already. 20160915 14:29:09< celticminstrel> I think I might still have an INSTALL update stashed somewhere actually. 20160915 14:29:46< celticminstrel> Main reason I didn't commit is because I haven't the slightest clue what the minimum requirement is for any library (mainly Boost) or how to determine this. 20160915 14:31:20< gfgtdf> celticminstrel: afaik 1.48 is the minimum required boost verison. 20160915 14:32:29< celticminstrel> Not 1.58? 20160915 14:32:44< celticminstrel> Pretty sure 1.54 was being used by myself and mattsc until recently though. 20160915 14:32:54< celticminstrel> Then again, that's no longer the case as far as I know. 20160915 14:34:35< mattsc> I’m using 1.59 now; yes, used to use 1.54, but I don’t remember how recently that was (on the order of half a year maybe?). 20160915 14:34:50< mattsc> could be more 20160915 14:34:52< celticminstrel> Definitely pre-C++11. 20160915 14:35:02< celticminstrel> For me at least. 20160915 14:35:10< celticminstrel> That is, before Wesnoth allowed C++11. 20160915 14:36:05< gfgtdf> celticminstrel: hmm no accoring to scronstript we even support 1.44 ut ir ecently rmeves all pre boost 1.48 wokraounds so its no moslikeley 1.48 20160915 14:41:26< gfgtdf> celticminstrel: i want to make the gamemap use a non-jagged map, you sure that you dont rmember aht your problems with boost multi array were? 20160915 14:42:53< celticminstrel> I'll check if it's recorded in a commit message, but otherwise, no I don't remember, though I think it was issues with getting it to compile. 20160915 14:44:15< celticminstrel> From what I understand, boost::multi_array supports dynamically changing not only the size but also the dimensions of the array; since I only needed to change the size, not the dimensions, I ended up writing a fairly simple vector2d instead. 20160915 14:44:42< celticminstrel> I seem to recall reading that multi_array also suports things like 1-based indexing. 20160915 14:45:00< celticminstrel> Though I might be confusing it with a different Boost class, I suppose. 20160915 14:48:08< celticminstrel> vultraz: Why do you always put the newlines in a string by themselves? 20160915 14:48:28< celticminstrel> "\n" and "\n" will work just fine. 20160915 14:48:37< vultraz> easier to identify 20160915 14:48:41< vultraz> at a glance 20160915 14:48:44< celticminstrel> I don't see how. 20160915 14:48:49 * vultraz shrugs 20160915 14:48:53< vultraz> it's just my style 20160915 14:48:57< celticminstrel> Though if you really think so you could at least consider omitting the << 20160915 14:49:14< celticminstrel> It's not really a big deal, but it's weird. 20160915 14:49:21< celticminstrel> (Both ways.) 20160915 14:49:29< gfgtdf> celticminstrel: wml supports escape chaters like \n ? 20160915 14:49:35< vultraz> and you like sticking large blocks of code together without newlines :P 20160915 14:49:40< vultraz> which is also weird 20160915 14:49:43< celticminstrel> gfgtdf: In C++ 20160915 14:49:48< celticminstrel> vultraz: It's not weird. 20160915 14:50:06< celticminstrel> I do put in blank lines in some places, though. 20160915 14:50:30< gfgtdf> celticminstrel: hmm i thought wml sinc eyou mentioned << escaping. 20160915 14:50:45< celticminstrel> gfgtdf: << insertion operator. 20160915 14:53:04< celticminstrel> gfgtdf: Sorry, my commit message doesn't mention what the issues with multi_array were. 20160915 14:53:09< vultraz> for the record gui2 wml does respect \n 20160915 14:53:13< vultraz> in strings 20160915 14:53:21< celticminstrel> Only GUI2 WML? 20160915 14:53:32< vultraz> actually all wml, yes 20160915 14:53:45< celticminstrel> So which is it? 20160915 14:54:04< vultraz> i guess all wml with text strings where they could be renderered 20160915 14:54:29< celticminstrel> Sounds like it's not specifically WML honouring it. 20160915 14:55:14< celticminstrel> (Which would mean it no longer contains \n once it's been loaded into a config object.) 20160915 14:55:39< irker112> wesnoth: Bär Halberkamp wesnoth:master 1acd76e9fe4f / data/campaigns/ (18 files in 6 dirs): Run wmlindent on data/campaigns https://github.com/wesnoth/wesnoth/commit/1acd76e9fe4fbbd359589dace3623dd5d5e8a783 20160915 14:57:03< celticminstrel> Elvish_Hunter: Any chance of updating wmlindent to not do this? https://github.com/wesnoth/wesnoth/commit/1acd76e9fe4fbbd359589dace3623dd5d5e8a783#diff-401dacfdd4a3f93e967162ccbbc973e2R684 20160915 14:57:28< celticminstrel> Basically, I think it should indent + continued lines an extra level, or similar. 20160915 14:58:34< celticminstrel> ...why does wmlindent remove blank lines? https://github.com/wesnoth/wesnoth/commit/1acd76e9fe4fbbd359589dace3623dd5d5e8a783#diff-a3c9029d3f4bce040fe90cee7db69ceeL415 20160915 14:58:46< celticminstrel> Mind you, I'd probably remove that particular one too, though. 20160915 14:59:25< celticminstrel> Here's another place that could maybe use work: https://github.com/wesnoth/wesnoth/commit/1acd76e9fe4fbbd359589dace3623dd5d5e8a783#diff-aa15a4cf284a9e547f5faa6d4f5dc3faR141 20160915 15:04:01< gfgtdf> celticminstrel: i think the last case should've just added a newline near the ( and ) 20160915 15:05:18< celticminstrel> gfgtdf: So {NEED_BARAN (x,y=31,21facing=se)} ? 20160915 15:05:38< gfgtdf> celticminstrel: yes 20160915 15:05:39 * celticminstrel using Python-style tokenization there. >_> 20160915 15:06:03< celticminstrel> Yeah, that makes sense, though I wonder if wmlindent is really equipped to deal with something that complicated. 20160915 15:06:48< gfgtdf> celticminstrel: i think wmlindend can handle that. fi te newlines were inserted. 20160915 15:07:15< celticminstrel> I meant more that I'm not sure if wmlindent is equipped to handle the newline insertion. 20160915 15:07:28< celticminstrel> But yeah, if someone wants to insert the newlines manually, then sure. 20160915 15:08:20-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc] 20160915 15:14:35< bumbadadabum> celticminstrel: I doubt it'd be hard to add it to wmlindent 20160915 15:14:47< bumbadadabum> idk about the macro case 20160915 15:14:53< bumbadadabum> since I think it reads the file line by line 20160915 15:15:06< bumbadadabum> but for the multiline strings it could be fixed 20160915 15:15:32< bumbadadabum> should be as easy as just making a case for + at the end of a line 20160915 15:15:37< bumbadadabum> ... or + at the start of one 20160915 15:16:05< celticminstrel> Huh? I thought it only worked if the + was at the end of the line? 20160915 15:16:22< celticminstrel> I also thought it worked if you have blah="some text" + # some comment 20160915 15:16:50-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 15:16:51< travis-ci> wesnoth/wesnoth#10956 (master - 1acd76e : Bär Halberkamp): The build is still failing. 20160915 15:16:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160190057 20160915 15:16:51-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 15:17:24< bumbadadabum> celticminstrel: wait is that in right now? 20160915 15:17:36< celticminstrel> bumbadadabum: What? 20160915 15:18:07< bumbadadabum> Huh? I thought it only worked if the + was at the end of the line? 20160915 15:18:46< celticminstrel> String continuation. I thought it only worked if the + was at the end of the line. I've never tried a + at the beginning of the following line though. 20160915 15:19:30< bumbadadabum> oh well idk about WML honestly 20160915 15:19:43< irker112> wesnoth: Celtic Minstrel wesnoth:master a3609e428b21 / data/campaigns/Two_Brothers/scenarios/01_Rooting_Out_a_Mage.cfg: TB: Minor formatting fixup https://github.com/wesnoth/wesnoth/commit/a3609e428b21e65c7e2c65bccbe2719db01d7d2d 20160915 15:19:48< bumbadadabum> In basically every language it WORKS on a new line 20160915 15:19:52< bumbadadabum> but you shouldn't be doing it 20160915 15:20:05< celticminstrel> I don't think that's really true in general? 20160915 15:20:27< celticminstrel> In languages where whitespace is not significant, sure. 20160915 15:20:51< bumbadadabum> is it significant in WML? 20160915 15:20:52< celticminstrel> But whitespace is significant in WML if I'm not mistaken. (Or at least, newlines are.) 20160915 15:21:14< bumbadadabum> yeah ok fair enough 20160915 15:21:16< celticminstrel> "x=4 y=5" is not the same as "x,y=4,5" or "x=4y=5" 20160915 15:21:25< bumbadadabum> newlines yeah 20160915 15:21:40< bumbadadabum> I think I'll try making a case for lines ending in + 20160915 15:21:45< bumbadadabum> or "+ #comment" 20160915 15:21:55< celticminstrel> Ah, you're gonna tackle the wmlindent case? Sure, why not. 20160915 15:22:02< celticminstrel> Assuming you're comfortable in Python. 20160915 15:22:06< celticminstrel> Python 3, even. 20160915 15:31:46-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20160915 15:32:32< bumbadadabum> celticminstrel: only thing I can't think about is how to deal with [set_variable] ... value = $var + 20160915 15:32:35< bumbadadabum> like that's a mistake 20160915 15:32:48< bumbadadabum> but will see the next line as a continued string 20160915 15:33:00< bumbadadabum> actually, I can think of something 20160915 15:35:07< celticminstrel> I kinda wish we could change the WML string concatenator to something else. Maybe & 20160915 15:36:56< bumbadadabum> celticminstrel: yeah, although that's a very uncommon thing 20160915 15:37:00< bumbadadabum> + is widely accepter 20160915 15:37:03< bumbadadabum> *accepted 20160915 15:37:11< celticminstrel> + is common in many languages 20160915 15:37:15< celticminstrel> Several use . or .. 20160915 15:37:21< celticminstrel> A few use & or && 20160915 15:37:48< celticminstrel> I'm only aware of Hypertalk (extinct) and AppleScript for the latter though. 20160915 15:39:54-!- Jetrel_ [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin] 20160915 15:40:08-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev 20160915 15:40:12-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Client Quit] 20160915 15:40:43-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 15:40:44< travis-ci> wesnoth/wesnoth#10957 (master - a3609e4 : Celtic Minstrel): The build is still failing. 20160915 15:40:44< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160197356 20160915 15:40:44-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 15:40:49-!- Jetrel [~Jetrel@2001:558:6014:1e:2422:435:dd84:bbf3] has joined #wesnoth-dev 20160915 15:41:55-!- Jetrel [~Jetrel@2001:558:6014:1e:2422:435:dd84:bbf3] has quit [Client Quit] 20160915 15:43:40-!- atarocch [~atarocch@62.134.204.4] has quit [Ping timeout: 260 seconds] 20160915 15:45:05-!- Jetrel [~Jetrel@2001:558:6014:1e:2422:435:dd84:bbf3] has joined #wesnoth-dev 20160915 15:48:56-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev 20160915 15:49:55-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 248 seconds] 20160915 15:50:05-!- Kwandulin [~Miranda@p200300760F2C71D338B83A61803763F4.dip0.t-ipconnect.de] has joined #wesnoth-dev 20160915 15:50:42-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20160915 15:51:31-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds] 20160915 15:51:31-!- wedge010 is now known as wedge009 20160915 15:52:19-!- prkc [~prkc@46.166.190.231] has joined #wesnoth-dev 20160915 15:53:02< bumbadadabum> celticminstrel: there are a few cases that MIGHT be undesirable with the change 20160915 15:53:13< celticminstrel> ? 20160915 15:53:39< bumbadadabum> message= _ "For this tutorial, you are playing Li’sar. You are standing on the keep, and your mentor Delfador is on the east side of the river. you can move the mouse over a unit to see a summary of its abilities and stats on the right of the screen." + {CONTINUE_MSG} 20160915 15:53:51< bumbadadabum> it adds another indent level here 20160915 15:55:02< celticminstrel> I dunno, it doesn't seem like a huge problem. 20160915 15:55:09< bumbadadabum> [note] 20160915 15:55:09< bumbadadabum> red,green,blue=255,255,192 20160915 15:55:09< bumbadadabum> description= _ "20% of finishing gold carried over to the next scenario." + " 20160915 15:55:09< bumbadadabum> "+ 20160915 15:55:09< bumbadadabum> #textdomain wesnoth 20160915 15:55:10< bumbadadabum> _"Notes:"+"" 20160915 15:55:11< bumbadadabum> #textdomain wesnoth-trow 20160915 15:55:13< bumbadadabum> [/note] 20160915 15:55:15< bumbadadabum> also, this in TRoW S15 20160915 15:55:53-!- JyrkiVesterinen [~JyrkiVest@89-166-122-141.bb.dnainternet.fi] has joined #wesnoth-dev 20160915 15:56:21< bumbadadabum> I could make a special case for #textdomain 20160915 15:56:33< celticminstrel> I can't really see what's going on there with my buggy client, but I guess it's the textdomain getting indented and then the following string not getting indented? 20160915 15:57:52< bumbadadabum> yes 20160915 15:58:07< celticminstrel> Yeah, that one seems like a problem. 20160915 15:58:33< celticminstrel> It's fine for the textdomain to be indented in my opinion, but the following string definitely needs to be too. 20160915 15:58:40< gfgtdf> did i oersee something or does boost multi_array not have a move ctor ? 20160915 15:58:44< gfgtdf> overlook* 20160915 15:59:03< bumbadadabum> I'll just add "or transform.startswith("#textdomain") 20160915 16:00:40< bumbadadabum> or any preprocessor stuff I guess 20160915 16:00:52< bumbadadabum> I mean, you SHOULD never be defining a macro at the start of a string 20160915 16:01:03< celticminstrel> textdomain is the only one that really matters, I think. Maybe regular comments. 20160915 16:01:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 16:01:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 16:01:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 16:02:45< bumbadadabum> yeah this fixes it 20160915 16:02:52< bumbadadabum> if there are any other issues I'll fix those later 20160915 16:04:45< irker112> wesnoth: Bär Halberkamp wesnoth:master b20cb7c413b9 / data/tools/wmlindent: wmlindent: Indent concatenated strings https://github.com/wesnoth/wesnoth/commit/b20cb7c413b92ec6141d8e8c79be0c7d73512bee 20160915 16:04:48< irker112> wesnoth: Bär Halberkamp wesnoth:master ce05a573cfba / data/ (13 files in 6 dirs): Run wmlindent https://github.com/wesnoth/wesnoth/commit/ce05a573cfba25e300b51110937ecaa791304fcf 20160915 16:05:58-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev 20160915 16:20:35< zookeeper> bumbadadabum, why change turn 10 to turn 10 refresh? 20160915 16:20:51< bumbadadabum> zookeeper: because Dacyn got rest healing the turn he reappears 20160915 16:21:53< zookeeper> okay, but presumably he also provided healing which was a nice surprise, and now he doesn't? 20160915 16:22:28< bumbadadabum> I guess healing that one turn right next to your original keep COULD be nice 20160915 16:22:47< bumbadadabum> but I found it never coming into play when I played through the scenario 20160915 16:23:05< bumbadadabum> if you want I can revert that 20160915 16:23:12< bumbadadabum> but I found the rest healing to look a little silly 20160915 16:23:27< bumbadadabum> maybe specify he shouldn't rest heal instead 20160915 16:23:43-!- JyrkiVesterinen [~JyrkiVest@89-166-122-141.bb.dnainternet.fi] has quit [Quit: Rebooting] 20160915 16:24:29< zookeeper> yes, you should be able to avoid rest healing by just setting .resting=no i think. 20160915 16:25:12< bumbadadabum> didn't put much thought into the healing 20160915 16:26:26< zookeeper> also the tutorial like you just quoted has missing capitalization 20160915 16:26:48-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 16:26:49< travis-ci> wesnoth/wesnoth#10958 (master - ce05a57 : Bär Halberkamp): The build is still failing. 20160915 16:26:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160208542 20160915 16:26:49-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 16:27:00< bumbadadabum> zookeeper: oh yeah it does 20160915 16:28:00-!- JyrkiVesterinen [~JyrkiVest@89-166-122-141.bb.dnainternet.fi] has joined #wesnoth-dev 20160915 16:32:39< zookeeper> huh, someone should really update those TRoW:15 carryover/note strings... 20160915 16:32:57< zookeeper> or wait, should they. 20160915 16:32:59 * zookeeper thinks 20160915 16:36:23< irker112> wesnoth: Bär Halberkamp wesnoth:master f1d6a96c1561 / data/campaigns/tutorial/scenarios/01_Tutorial_part_1.cfg: Tutorial: Fix a missing capital in S1 https://github.com/wesnoth/wesnoth/commit/f1d6a96c1561a9f218e9e6fe67bbb2052bf4efaf 20160915 16:36:26< irker112> wesnoth: Bär Halberkamp wesnoth:master 7d5f35dd68a6 / data/campaigns/Eastern_Invasion/scenarios/01_The_Outpost.cfg: EI S1: Make Dacyn able to heal others when reappearing https://github.com/wesnoth/wesnoth/commit/7d5f35dd68a6a4942860e4cfb0b3ac22bca195d7 20160915 16:40:48< zookeeper> well, looks like there's not really anything obsolete in those objectives after all. just missing a newline in one spot. 20160915 16:41:22< bumbadadabum> so I decided to try out my changes to ToD lighting from 2013 20160915 16:41:27< bumbadadabum> that I never committed 20160915 16:43:53< bumbadadabum> what would be a good way to get comments on changes to ToD RGB values? 20160915 16:44:21< zookeeper> make a PR 20160915 16:44:27< bumbadadabum> good point 20160915 16:44:39< zookeeper> and in fact, please make a PR for any non-trivial campaign changes you might be planning in the future 20160915 16:44:54-!- boucman_work [~boucman@193.56.60.161] has quit [Remote host closed the connection] 20160915 16:45:24-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity] 20160915 16:45:31-!- atarocch [~atarocch@HSI-KBW-46-223-95-173.hsi.kabel-badenwuerttemberg.de] has joined #wesnoth-dev 20160915 16:46:22-!- louis94 [~~louis94@91.178.241.97] has joined #wesnoth-dev 20160915 16:46:44-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev 20160915 16:48:45< bumbadadabum> zookeeper: were my EI changes that bad? 20160915 16:50:03< celticminstrel> I doubt it's a question of whether they were bad. 20160915 16:50:55< bumbadadabum> idk I've just never had shit about just pushing to master 20160915 16:51:24< zookeeper> if you count lots of unexplained weird changes some of which break stuff as bad, then... well... 20160915 16:51:56< bumbadadabum> what kind of change then? 20160915 16:52:04< bumbadadabum> I don't recall breaking anything per se 20160915 16:52:17< bumbadadabum> I mostly had trouble with the fucked SP/MP merger at the time 20160915 16:52:22< bumbadadabum> which caused carryover to be broken 20160915 16:52:32< bumbadadabum> ... which explains some of the weird carryover numbers in EI 20160915 16:52:50< celticminstrel> Why does team::calculate_enemies() return bool... 20160915 16:53:14< zookeeper> how does the SP/MP merger explain weird carryover numbers? 20160915 16:54:04< bumbadadabum> zookeeper: it didn't work properly during the time I worked on EI 20160915 16:54:07< bumbadadabum> IIRC 20160915 16:54:23< zookeeper> yes, that's not what i asked 20160915 16:54:37< bumbadadabum> I have no clue but then again I don't completely remember 20160915 16:54:42< zookeeper> okay 20160915 16:54:49< bumbadadabum> anyway idk what other things I broke 20160915 16:54:59< bumbadadabum> since I actually do try to not fuck up constantly but it's hard ;_; 20160915 16:55:33-!- Kwandulin [~Miranda@p200300760F2C71D338B83A61803763F4.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 20160915 16:55:50-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 16:55:51< travis-ci> wesnoth/wesnoth#10959 (master - 7d5f35d : Bär Halberkamp): The build is still failing. 20160915 16:55:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160216341 20160915 16:55:51-!- travis-ci [~travis-ci@ec2-54-204-241-243.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 16:56:32-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: http://www.kvirc.net/ 4.9.2 Aria] 20160915 16:56:51-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20160915 16:58:09< zookeeper> the bandits code was broken (might have worked on the version you used though), the knight hostage thing was buggy, nonsensical changes like removing an amulet, making a lich an ancient lich, the carryover numbers, owaec being loyal as a leader... but i haven't liked analyzed every commit 20160915 16:58:21< zookeeper> s/liked/like 20160915 16:59:39< bumbadadabum> Oh the knight thing I was going to fix but got frustrated with wesnoth 20160915 16:59:46< bumbadadabum> I tested the bandits code at the time 20160915 16:59:56< bumbadadabum> I don't know where I removed an amulet 20160915 17:00:21< zookeeper> in c8adbf18c5 i believe 20160915 17:00:37< bumbadadabum> and idr where I made a lich an ancient lich 20160915 17:01:56< bumbadadabum> zookeeper: ahh I remember that now 20160915 17:02:21< zookeeper> well, i trust you see why i'd prefer to sanity-check stuff beforehand 20160915 17:02:31< bumbadadabum> I was going to place it somewhere else after playtesting the scenario... but then I stopped and gave up 20160915 17:02:39< bumbadadabum> Yeah I see now 20160915 17:02:50< bumbadadabum> That's honestly quite unprofessional sorry 20160915 17:03:11-!- Kwandulin [~Miranda@p200300760F2C71D365F4202C098EC667.dip0.t-ipconnect.de] has joined #wesnoth-dev 20160915 17:03:58-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 17:04:37< irker112> wesnoth: ln-zookeeper wesnoth:master f88cf93353e8 / data/campaigns/The_Rise_Of_Wesnoth/scenarios/15_A_New_Land.cfg: Added a newline to space apart gold carryover and notes sections https://github.com/wesnoth/wesnoth/commit/f88cf93353e819ab446aca96286be41fa1f9e05f 20160915 17:05:33< bumbadadabum> zookeeper: I've been trying to get better at not being retarded but I'll PR things for now 20160915 17:06:43< celticminstrel> In particular, PR anything that a) you're uncertain about or b) is large and would benefit from review. 20160915 17:07:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds] 20160915 17:07:08-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 17:07:12< zookeeper> cool 20160915 17:07:27< bumbadadabum> celticminstrel: yes I know 20160915 17:11:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 17:15:59-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20160915 17:15:59-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20160915 17:16:58-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20160915 17:18:17-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev 20160915 17:18:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 17:22:25< celticminstrel> JyrkiVesterinen: There are a lot of C4018 (signed/unsigned mismatch) and C4804 (seems to be for <= etc on bool), do you know if these can be fixed? If not, maybe they should be suppressed instead? 20160915 17:23:16< JyrkiVesterinen> I believe they are fixable. 20160915 17:25:03< gfgtdf> celticminstrel: on current master? coudl you give an example line ? 20160915 17:25:23< celticminstrel> team.hpp:243 20160915 17:25:38< celticminstrel> (That's the only one in my log right now since I rebuild just a few files) 20160915 17:25:49-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 17:25:50< travis-ci> wesnoth/wesnoth#10961 (master - f88cf93 : ln-zookeeper): The build is still failing. 20160915 17:25:50< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160222918 20160915 17:25:50-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 17:26:57< celticminstrel> ^rebuilt 20160915 17:27:11< gfgtdf> celticminstrel: hmm thats 'return enemies_[index];' here. 20160915 17:28:23< celticminstrel> It's possible my line numbers are slightly different in that file. 20160915 17:28:54< celticminstrel> Ah, that one was a bad example - it's an error introduced by me. 20160915 17:30:13< celticminstrel> Also some C4800 (implicitly casting int to bool) 20160915 17:31:24< gfgtdf> zookeeper: you know how [terrain_graphics] works ? 20160915 17:31:28< zookeeper> yes 20160915 17:31:47< zookeeper> gfgtdf, saw my comment about the location range thing? 20160915 17:32:03< gfgtdf> zookeeper: i mena aving different sizes rows ? 20160915 17:32:07< gfgtdf> zookeeper: no i didnt saw 20160915 17:32:23< zookeeper> does anyone know why write_location_range si so cimplicated i'd think that it just does 'for loc in range do { x << ',' << loc.x+1; y << ',' << loc.y+1 }' but that doesn't seem ot be the case. 20160915 17:32:24< zookeeper> gfgtdf, well surely it's complicated so that it can write compact ranges instead of listing each location individually? like x=1-10 instead of x=1,2,3,... 20160915 17:32:35< gfgtdf> zookeeper: ah yes i read that one srty 20160915 17:32:48< gfgtdf> sry* 20160915 17:34:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 17:43:42< zookeeper> so what was that about [terrain_graphics]? 20160915 17:43:52< shadowm> vultraz: Can you point me to a particular line and file for one of those "excessive" line breaks/indents? 20160915 17:44:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 17:47:25-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: http://www.kvirc.net/ 4.9.2 Aria] 20160915 17:47:26-!- Nobun [~nobun@5.170.110.103] has joined #wesnoth-dev 20160915 17:49:34< Nobun> celticminstrel: sorry to say it, but the wmlxgettext development is suspended for a while... I will resume the work as soon as I can. No other new progress was done after the previous WIP informations 20160915 17:50:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 17:51:11-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20160915 17:52:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 17:52:55-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 17:53:47< celticminstrel> Nobun: Like I mentioned earlier, there's no hurry. You have something like 6 months before we really need it. 20160915 17:54:02< irker112> wesnoth: Jyrki Vesterinen wesnoth:master 42c3435c0c1d / projectfiles/VC12/ (4 files): Update Visual Studio project https://github.com/wesnoth/wesnoth/commit/42c3435c0c1d9e6a57949331f419542c8a3c71dd 20160915 17:54:04< irker112> wesnoth: Jyrki Vesterinen wesnoth:master dc89612dc4cd / src/server/server_base.cpp: Fix wesnothd build for Windows https://github.com/wesnoth/wesnoth/commit/dc89612dc4cde077dfa55dd062925bffdc77f6c8 20160915 17:54:30< celticminstrel> shadowm: Only think I noticed was indenting conditional includes. Dunno if that's what vultraz is referring to though. 20160915 17:56:19< Nobun> Yes, celticminstrel: but I think it is my duty to inform you about progress and problems... so it should not happen you know about problems only at the very last moments 20160915 17:56:29< gfgtdf> zookeeper: sry, my question was whether the map= attribute in [terrain_graphics] allows rows of diferent size 20160915 17:57:17< celmin> Sure. 20160915 17:57:56< zookeeper> gfgtdf, hmh... i don't think so. you probably need to use . as padding if you want to make a non-rectangular map. if that's what you meant. 20160915 17:58:44< gfgtdf> zookeeper: ok thx 20160915 17:58:52< tad_> gfgtdf: Definetly not. All rows must be same size. There is a specific check that the map is a rectangle 20160915 17:59:08< gfgtdf> tad_: where? 20160915 17:59:17< zookeeper> he's not talking about maps, but a terrain graphics thing 20160915 17:59:50< tad_> gfgtdf: I think it was a comment in map.cpp where it's loading the file/string into the class 20160915 18:00:23< tad_> gfgtdf: sorry, misread. my mind is on terrain_mask 20160915 18:02:17< tad_> zookeeper: If I give ambush to an Elvish Archer and it advances to Elvish Ranger, does the ambush duplicate? 20160915 18:02:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 18:04:58< zookeeper> tad_, uh... maybe :J 20160915 18:05:09< zookeeper> i don't know if there's a check against duplicates (based on id) if not 20160915 18:05:11< zookeeper> or not 20160915 18:05:30< tad_> Well, I'm testing it. Crossing fingers it doesn't blow something up. 20160915 18:06:03-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Ping timeout: 240 seconds] 20160915 18:06:07< Nobun> gtgtdf: also in other games maps has a regular rows and columns (hexagonal or ortogonal, don't matter). If you want an irregular size map (like a labyrint) the best strategy, imho, is to use impassable void tile (it is also what you do in other games when mapping) 20160915 18:06:56< Nobun> introducing different rows and colums in a different way can introduce bugs and problems, imho 20160915 18:09:09< gfgtdf> tad_: i have already seen unit its woudl swapon specials multiple times in UMC, so i woudlnt be surpised if the dubblicate in that case. 20160915 18:10:10< zookeeper> tad_, but in any case that'd be easy to guard against by [effect] [filter] 20160915 18:11:04< celmin> tad_: Shouldn't 20160915 18:11:18< tad_> Does not appear to duplicate. Shows once on side panel, and only once on recruit list. Unlike that mess with leadership (which is not duplicates). 20160915 18:12:15< tad_> Is this new? Console message: "window key handler fires" 20160915 18:13:56-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 18:13:57< travis-ci> wesnoth/wesnoth#10963 (master - dc89612 : Jyrki Vesterinen): The build is still failing. 20160915 18:13:57< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160238463 20160915 18:13:57-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 18:16:50 * tad_ chuckles. When testing it's best to keep moves and hitpoints straight. hitpoints=999 makes it sorta hard to quick-kill using a normal attack 20160915 18:17:02< celmin> Heh 20160915 18:19:59< bumbadadabum> celmin: since you mentioned trying my PR, I didn't update the 24-hour or post-fall schedule yet so if you need a comparison... 20160915 18:20:19< celmin> 'kay 20160915 18:23:09< tad_> gui/widgets/window.cpp: std::cerr << "window key handler fires\n"; 20160915 18:23:16< tad_> Is that temporary? 20160915 18:25:05-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev 20160915 18:25:08< celmin> No idea! Might be@ 20160915 18:25:09< celmin> ! 20160915 18:25:18< mordante> servus 20160915 18:27:36< tad_> I need to make a PR for git .. 'git kick-knee' similar to 'git blame' but automatically sends an email to the guilty party 20160915 18:28:12< celmin> XD 20160915 18:28:32< celmin> Though that won't work for people who use fake emails. 20160915 18:31:15< celmin> I don't think I would recommend aquileia's repo to anyone, at least not with respect to Boost. 20160915 18:31:29< celmin> I'm sure it's handy for glib and co. 20160915 18:32:06< celmin> But not including the entirety of Boost seems like a bad idea. 20160915 18:34:45< tad_> ? 20160915 18:35:37< tad_> zookeeper: I believe I'm done with the requests and comments for the HttT PRs. 20160915 18:37:00-!- Nobun [~nobun@5.170.110.103] has quit [Ping timeout: 244 seconds] 20160915 18:38:49< celmin> Enumerating the GUI1 list again… help, stats, attack prediction, in-game UI… was there something else still? 20160915 18:38:53< celmin> vultraz: ^ 20160915 18:39:01< celmin> Oh right, MP Connect. 20160915 18:39:02-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev 20160915 18:42:57 * tad_ just LOVES Boost. One little typo and you get one 'note' .. not even a warning .. and it overflows his 80x200 console. How in the world are you supposed to make heads or tails out of an error message which is larger than most source files? 20160915 18:44:35< JyrkiVesterinen> It is also nice that boost::bimap ends up generating type names which are so long that MSVC truncates them. :D 20160915 18:44:49< JyrkiVesterinen> Celmin turned off the respective compiler warning last week. 20160915 18:45:15-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds] 20160915 18:45:55< tad_> There's a blog I read somewhere over the past week ranting about how C++ programmers don't know how to program in C++. Junk like that is the proof. 20160915 18:46:09< celmin> The problem I'm getting is that somehow I seem to have Boost 1.62 headers mixed in with my Boost 1.58 headers, based on the link errors I get. 20160915 18:46:57< celmin> Boost's auto-link system seems kind of annoying now. 20160915 18:47:21< mordante> c++ error message can indeed be challenging to read 20160915 18:47:27< tad_> sounds like something didn't get rebuilt. 20160915 18:47:33< loonycyborg> that's what ides are for 20160915 18:47:34< celmin> tad_: It's not that. 20160915 18:47:49< mordante> some of the boost libraries are able to produce multiple MBs of erros 20160915 18:47:54< mordante> errors* 20160915 18:48:05< loonycyborg> without ide I have to redirect compiler output to text editor 20160915 18:48:07< celmin> Boost headers that have a corresponding lib automatically include "#pragma comment" lines to ensure that the lib gets linked in MSVC. 20160915 18:48:18< loonycyborg> like "compiler_command | gvim -" 20160915 18:48:42< loonycyborg> qt creator at least displays them separated into items 20160915 18:48:44< tad_> errors like that are best done with >/dev/null 20160915 18:48:57< loonycyborg> put pretty sure this can be improved 20160915 18:49:09< loonycyborg> like having long type names being collapsed 20160915 18:49:18< loonycyborg> and expanded with a gui control 20160915 18:49:43< celmin> XCode groups all notes under the preceding error/warning. 20160915 18:50:07< celmin> MSVC on the other hand doesn't even bother to include the notes (you need to view the raw log for them, when they exist). 20160915 18:50:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 18:50:14< tad_> Or one could encapsulate so you only see those insane messages when you're working on the object itself and your programmer/users see a nice, clear message when they use your object incorrect.y 20160915 18:50:50< celmin> That would be nice. I think that was the intent behind things like static_assert. 20160915 18:51:00< loonycyborg> that's what c++ concepts are supposed to help with 20160915 18:51:05< celmin> That too. 20160915 18:51:29< tad_> Too bad C++ programers so rarely use C++ 20160915 18:52:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!] 20160915 18:53:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 18:53:13-!- louis94 [~~louis94@91.178.241.97] has quit [Ping timeout: 272 seconds] 20160915 18:53:15< shadowm> Checking for Boost smart_ptr library... yes 20160915 18:53:32< shadowm> TIL scons tests for the presence of very specific Boost header-only libraries. 20160915 18:53:44-!- Nobun [~nobun@5.170.107.37] has joined #wesnoth-dev 20160915 18:54:02< celmin> tad_: What? 20160915 18:54:02< loonycyborg> I'm not the one who added those tests iirc 20160915 18:55:43< tad_> celmin: Oh, just ranting a bit. Look at PR 775. Only read the three lines of code I added. Can YOU tell what they're doing? What those variables mean and what's changing? 20160915 18:57:22-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev 20160915 18:57:56< celmin> tad_: 775? The terrain_mask one? 20160915 18:58:03 * tad_ nods. 20160915 18:58:07< shadowm> loonycyborg: The production 1.13 and trunk campaignd instances are running with Asio now. 20160915 18:58:19< loonycyborg> cool 20160915 18:58:27< celmin> I don't see any "three lines of code added"… more like five or so lines rewritten... 20160915 18:58:55< shadowm> I'll toss at you any surprising bugs that crop up. 20160915 18:59:04< loonycyborg> ok 20160915 18:59:07< shadowm> Hopefully less than last time with wesnothd. 20160915 18:59:12< tad_> starting_positions_.left.erase(pair.first); 20160915 18:59:36< tad_> left? pair? first? 20160915 18:59:45< celmin> Ah. 20160915 18:59:49< loonycyborg> seems like boost.bimap 20160915 18:59:56< tad_> It is 20160915 18:59:58< celmin> Yeah, I find the bimap thing a bit weird. 20160915 19:00:26< celmin> Noticed that when in wesnoth.special_locations implementation, too. 20160915 19:00:31< celmin> ^-when 20160915 19:00:36< loonycyborg> I used it in wesnothd at first too, but later switched to using full multi_index 20160915 19:01:01< loonycyborg> at least it supports explicitly naming indices with tags 20160915 19:01:15< loonycyborg> instead of using cryptic stuff like left and right 20160915 19:03:09< tad_> Who cares how 'special_positions' is implemented? bimap? SQL database? NoSQL table? What matters is you have it, it has certain fields and you can do certain operations. But no .. it's so married to bimap fixing it now will be a nightmare. 20160915 19:03:48< tad_> And that's just one of .. what .. hundreds, thousands ?? .. of places in wesnoth 20160915 19:05:23-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving] 20160915 19:05:29 * shadowm blinks. 20160915 19:06:35-!- ancestral [~ancestral@173-8-126-226-Minnesota.hfc.comcastbusiness.net] has joined #wesnoth-dev 20160915 19:06:42< irker112> wesnoth: Celtic Minstrel wesnoth:dynamic_bitset fe08701b7ca3 / src/ (27 files in 13 dirs): Change most vector to dynamic_bitset https://github.com/wesnoth/wesnoth/commit/fe08701b7ca3639b17dca20023d2bb34433042c6 20160915 19:08:01< celmin> ^ Please review (PR 781) 20160915 19:08:13< gfgtdf> celmin: what are 5teh advantage sof dynamic_bitset ? 20160915 19:09:23< celmin> Main one is that it provides an interface more similar to std::bitset, in particular with any(), all(), none(). 20160915 19:09:48< celmin> Also, it won't suddenly take 8x the space if the C++ standard decides to remove the vector specialization some day. 20160915 19:11:48< celmin> It has the minor downside of not working with standard algorithms and range-for, but I don't think that's a problem - there are no or almost no cases of iterating over a vector 20160915 19:12:03< celmin> (There were a couple of uses of iterators and algorithms which were easily replaced.) 20160915 19:13:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20160915 19:16:21< tad_> I get so tired of waiting for wesnoth to recompile 300 files when all I want is to check one. 20160915 19:21:56-!- JyrkiVesterinen [~JyrkiVest@89-166-122-141.bb.dnainternet.fi] has quit [Quit: .] 20160915 19:25:22< gfgtdf> celmin: you rmoved the parmeter form states_ in unit.cpp why is that ? 20160915 19:26:51< celticminstrel> It was using vector as a fixed-size bitset, so I changed it to std::bitset, for which default construction gives the same effect as those parameters did on vector - ie, filled with all zeros. 20160915 19:29:08< gfgtdf> celmin: pr looks fine to me then. 20160915 19:29:40< celticminstrel> I'm assuming that means that you've concluded there are no (accidental) functional changes. 20160915 19:33:17< gfgtdf> celmin: hmm yes but i only loooked at the diff. I think the main 'risk' here is that there coudl be functions that have the same name for vector and danamic_bitset but diferent meaning that are used in non-changed code. 20160915 19:33:24< zookeeper> vultraz, so why does haldric show up as an un-TC'd "?" in TRoW saves in the new load dialog? 20160915 19:33:44< celticminstrel> I think that risk is fairly low... 20160915 19:34:06< celticminstrel> I don't think there are many functions that have the same name for vector and dynamic_bitset, anyway. 20160915 19:34:16< celticminstrel> I know push_back has the same behaviour in both. 20160915 19:36:09< tad_> zookeeper: That happens to all units which come from the campaign instead of core 20160915 19:36:24< tad_> zookeeper: ^unit_type's 20160915 19:37:42< celticminstrel> zookeeper: Something to do with the save index. 20160915 19:38:05< zookeeper> right, so already scheduled to be fixed then, good 20160915 19:38:09< celticminstrel> In an ideal world, you would only see that if you deleted your save index. 20160915 19:38:25< celticminstrel> (I don't think there's any easy way to make it show correctly if you've deleted your save index.) 20160915 19:39:54 * celticminstrel will wait for vultraz to okay the dynamic_bitset PR before merging it. 20160915 19:40:34-!- ancestral [~ancestral@173-8-126-226-Minnesota.hfc.comcastbusiness.net] has quit [Quit: i go nstuf kthxbai] 20160915 19:42:22< celticminstrel> In particular, I didn't remove the any_rows_shown() function. I figured it was actually marginally faster to keep it, since it doesn't need to walk the whole listbox. 20160915 19:42:46< celticminstrel> If the bitset were cached or something rather than constructed on request, that would be a different story. 20160915 19:47:46-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20160915 19:47:52-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20160915 19:48:22-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 19:48:23< travis-ci> gfgtdf/wesnoth-old#685 (location_refeactor - 060eb82 : gfgtdf): The build is still failing. 20160915 19:48:23< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/160264177 20160915 19:48:23-!- travis-ci [~travis-ci@ec2-54-242-123-125.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 19:59:40-!- gfgtdf [~chatzilla@x4e368915.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617]] 20160915 20:06:57-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed] 20160915 20:07:27-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 244 seconds] 20160915 20:13:24-!- Kwandulin [~Miranda@p200300760F2C71D365F4202C098EC667.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 20160915 20:13:50-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev 20160915 20:15:25-!- gfgtdf [~chatzilla@x4e368915.dyn.telefonica.de] has joined #wesnoth-dev 20160915 20:16:16< gfgtdf> zookeeper: is there any documentation about the the map= format in [scenario] ? https://wiki.wesnoth.org/ScenarioWML says 'See BuildingMaps for a description of the Wesnoth map syntax.' but BuildingMaps doesnt decribe the map syntax at all 20160915 20:17:26< zookeeper> gfgtdf, there's https://wiki.wesnoth.org/TerrainCodesWML#Terrain_codes_in_map_data 20160915 20:17:33-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Ping timeout: 240 seconds] 20160915 20:17:42< gfgtdf> zookeeper: ok thx 20160915 20:17:59-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds] 20160915 20:18:06< celticminstrel> gfgtdf: Note that the header (border_size etc) has been removed. 20160915 20:18:23-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev 20160915 20:19:00< gfgtdf> celticminstrel: yes but the c++ code is still there: https://github.com/wesnoth/wesnoth/blob/master/src/map/map.cpp#L239 and i wonder whether i shodul remove it 20160915 20:20:38< celticminstrel> Not sure. vultraz doesn't like the header and I know no reason why it would be brought back, though. 20160915 20:21:03< celticminstrel> It certainly wouldn't be brought back with the same function as before (specifying border-size and usage). 20160915 20:24:03-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds] 20160915 20:27:10-!- Nobun [~nobun@5.170.107.37] has quit [Ping timeout: 260 seconds] 20160915 20:28:57-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev 20160915 20:30:36-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20160915 20:31:12< zookeeper> as long as it'll still cope with old maps that have border_size... 20160915 20:38:13-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV] 20160915 20:45:47-!- Nobun [~nobun@5.170.108.83] has joined #wesnoth-dev 20160915 20:49:03-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20160915 20:51:01-!- Nobun [~nobun@5.170.108.83] has quit [Ping timeout: 272 seconds] 20160915 20:58:23< irker112> wesnoth: mattsc wesnoth:fix-tunnel-issues 045284896615 / src/ (actions/move.cpp pathfind/teleport.cpp): Check for occupied tunnel exits at teleport map creation https://github.com/wesnoth/wesnoth/commit/045284896615f20488ddb600a8c5b4c2da28e752 20160915 20:58:25< irker112> wesnoth: mattsc wesnoth:fix-tunnel-issues 692109f3bbcd / src/pathfind/ (teleport.cpp teleport.hpp): [tunnel] tag: add new optional parameter pass_allied_units=yes/no https://github.com/wesnoth/wesnoth/commit/692109f3bbcd0489d81f7f244c7bc244b703c669 20160915 20:58:32< gfgtdf> zookeeper: wiki says that only border_size=1 is allows for maps anyway. 20160915 20:59:30< mattsc> celmin, zookeeper, gfgtdf, DeFender1031, tad_, whoever is interested: that ^ is in a branch in PR 782: https://github.com/wesnoth/wesnoth/pull/782 20160915 21:00:12< gfgtdf> mattsc: does also also add an option see_though= ye/no ? 20160915 21:00:21< gfgtdf> mattsc: (or equivlant) 20160915 21:00:46 * celticminstrel will check that out on the other computer in a little while, to avoid having to launch Firefox. 20160915 21:00:52< mattsc> gfgtdf: no, it doesn’t. I’d prefer to fix the moves first; and deal with vision later separately. 20160915 21:01:07< mattsc> As in, in a separate PR. 20160915 21:01:47< gfgtdf> zookeeper: do you know which mainline campaigns uses [terrain_mask] that can be used as testcases? 20160915 21:04:45< DeFender1031> mattsc, nice 20160915 21:06:24< mattsc> DeFender1031: thanks; and I think now that I have this figured out (took me long enough!), I might be in a better position to look into some of the other things you suggested a few days ago. 20160915 21:06:56< DeFender1031> :D 20160915 21:10:06< zookeeper> gfgtdf, as testcase for what? 20160915 21:10:40< gfgtdf> zookeeper: that i dont break [terrain_mask] when refactoring mpa locations 20160915 21:16:07< zookeeper> gfgtdf, not many especially good+easy testcases in mainline i guess. 20160915 21:16:39< gfgtdf> zookeeper: ok 20160915 21:17:05< zookeeper> if you want to try some of them, then the ones in liberty, HttT and DW would probably be better than the others 20160915 21:19:55-!- louis94 [~~louis94@91.178.241.245] has joined #wesnoth-dev 20160915 21:20:32< gfgtdf> zookeeper: thx 20160915 21:23:34< gfgtdf> zookeeper: is it intended that most mainline maps that their masks inlined (instead ok mask={maps/mask.map} whihc woudl be easier readable by the wesnoth editor) ? 20160915 21:25:36-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 21:26:38-!- travis-ci [~travis-ci@ec2-54-211-188-111.compute-1.amazonaws.com] has joined #wesnoth-dev 20160915 21:26:39< travis-ci> wesnoth/wesnoth#10969 (fix-tunnel-issues - 692109f : mattsc): The build failed. 20160915 21:26:39< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160286663 20160915 21:26:40-!- travis-ci [~travis-ci@ec2-54-211-188-111.compute-1.amazonaws.com] has left #wesnoth-dev [] 20160915 21:27:11< zookeeper> gfgtdf, yeah, they're all really simple. except the one in liberty, which i guess one might want to split off into a separate file the next time it's edited. 20160915 21:30:03< zookeeper> tad_, don't you think "Air mobility" sounds a bit anachronistic? :P 20160915 21:30:28< zookeeper> err 20160915 21:30:32< zookeeper> wait what's the right word 20160915 21:31:25 * celticminstrel notes that the build has been failing since some credits thing, so that's probably not mattsc's fault. 20160915 21:31:28< tad_> Dunno. Spent most of my life near an air force base and heard it a lot. What I was going for was Li'sar is trained and would think first to use them tactically. I'm open to alternatives 20160915 21:31:29< zookeeper> oh yeah, anachronistic 20160915 21:32:00< tad_> mattsc: I've not used tunnels but from your comments, sounds like a good fix. 20160915 21:33:49< tad_> So .. anyone .. as a warrior trained in tactics .. upon first seeing air units (Gryphons) and thinking how you would use them .. what would your comment be? 20160915 21:35:24-!- gfgtdf [~chatzilla@x4e368915.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617]] 20160915 21:37:46< tad_> zookeeper: Does "Attacking from the air? ..." work as an alternative? 20160915 21:38:20< celmin> Slightly surprised this is still connected. 20160915 21:38:29< celmin> It tends to disconnect when left idle for awhile. 20160915 21:38:30< zookeeper> i dunno, i don't see why she has to comment on their tactical use 20160915 21:39:09< tad_> Well, I can take it out. She wasn't present for Gryphon Mountain and my thinking was 20160915 21:39:13 * celmin wonders if tad_ or zookeeper could link the specific line under contentions. 20160915 21:39:21-!- gfgtdf [~chatzilla@x4e368915.dyn.telefonica.de] has joined #wesnoth-dev 20160915 21:39:29< gfgtdf> tad_: you tested the honor special locations pr ? 20160915 21:39:35< tad_> (1) new to her and (2) she spend a lot of time at the war college iirc 20160915 21:40:24< tad_> gfgtdf: yes. Offset be even and odd and did a diagonal of grass then [unit] to the names on the grass to ensure the units appeared on the grass. had a bug as you expected and fixed it 20160915 21:41:36< tad_> celmin: see pr 737 [message]s on turn 4 added block of code 20160915 21:42:04< gfgtdf> tag_: sure that store_special_location= is needed ? there is already [store_locations]location_id= which shodul do the same 20160915 21:42:11 * celmin will take a few minutes to get to it, looking over mattsc's first. 20160915 21:42:40< celmin> gfgtdf: Let me guess, that was never documented? 20160915 21:42:41< tad_> gfgtdf: no. let me check the wiki. so may tags such old grey matter ... 20160915 21:42:59< gfgtdf> tad_: not sure whether its docuemneted in the wiki though 20160915 21:43:40< tad_> SLF does not mention id= 20160915 21:43:46< gfgtdf> celmin: well it was added quite recently and i didnt add it to the wiki yet beasuce it still wasnt 100% sure on the name. 20160915 21:44:07< celmin> Wait, why location_id rather than id? 20160915 21:44:36< celmin> gfgtdf: BTW, if you define a special location called "leader", is it impossible to place units there? 20160915 21:44:44< celmin> (using placement=) 20160915 21:44:49< zookeeper> tad_, well personally i'd first merge delfador's line into konrad's ("Such wondrous beasts! It appears our aid in their mountain homeland is being repaid.") 20160915 21:45:02< gfgtdf> celmin: yes. 20160915 21:45:11< tad_> zookeeper: OK. 20160915 21:45:21< celmin> That feels a little sloppy, but whatever... 20160915 21:45:24< zookeeper> tad_, i have no particular ideas for li'sar 20160915 21:45:27< gfgtdf> celmin: well you can sure [store_location] + action wml ofc 20160915 21:45:43 * celmin will take a look at Li'sar's line and see if I can come up with any ideas. 20160915 21:46:00< tad_> zookeeper: I'll go with that change, then. Take a moment to edit and push. Unless you want me to wait. 20160915 21:46:16-!- Sirp_ [~Sirp@u17402953.onlinehome-server.com] has quit [Ping timeout: 250 seconds] 20160915 21:46:22< celmin> It's not highly important, but if you wait I can see the original version. 20160915 21:46:44< tad_> celmin: OK. 20160915 21:47:08-!- aidanhs [~aidanhs@2a00:d880:6:1ad::8e27] has quit [Ping timeout: 250 seconds] 20160915 21:52:32-!- Appleman1234_ [~Appleman1@KD119104049219.au-net.ne.jp] has joined #wesnoth-dev 20160915 21:52:33-!- Appleman1234 [~Appleman1@KD119104112091.au-net.ne.jp] has quit [Ping timeout: 276 seconds] 20160915 21:52:40-!- Sirp [~Sirp@u17402953.onlinehome-server.com] has joined #wesnoth-dev 20160915 21:56:23< tad_> zookeeper, celmin I'm going to push the change without merging to the original commit for before/after comparison. If it's good I'll merge them and re-push 20160915 21:57:11< celmin> ;kay 20160915 21:57:13< celmin> 'kay 20160915 21:58:48< tad_> I like that Li'sar's comment points out you missed Gryphon Riders when you didn't get the eggs. 20160915 21:59:56< celmin> Hmmm… 20160915 22:00:24< celmin> Pffft, Li'sar mad. 20160915 22:00:31< tad_> celmin about SLF if store locations has id= support in some form and it's jsut not documented, I should pull the new tag from PR 775 20160915 22:00:46-!- aidanhs [~aidanhs@2a00:d880:6:1ad::8e27] has joined #wesnoth-dev 20160915 22:01:02< tad_> celmin: well image doesn't really look it to me. The sword out is what I went after. 20160915 22:01:31< celmin> I like it conceptually, but somehow it seems awkward. 20160915 22:03:06< zookeeper> tad_, i don't see how it points that out. she just wonders whether they could carry troops, which... gives the player the impression that that's going to be possible, except it won't be 20160915 22:03:21< celmin> This doesn't address it, but a reference to "if only we had the time" might be good. 20160915 22:04:41< tad_> Well, my normal reaction at this point is just drop it. Doesn't matter, really. It's just a message to Esc over ... 20160915 22:05:41< celmin> What I just said might however address what zookeeper just said. Maybe? 20160915 22:06:03< zookeeper> gryphon mountain happened ages ago, i don't think one needs to really make reference to how you missed something by doing things the way you did all the way back then. some kind of comment as to how they will be useful now is of course in order (although probably not necessary; their capabilities are probably rather obvious) 20160915 22:07:52< celmin> It's not strictly speaking a reference to what you missed, so much as a character moment speculating about possibilities… which happen by coincidence(?) to be something you missed. 20160915 22:08:28< zookeeper> but still, it's a bit of an easter egg in the first place which newbies are unlikely to stumble upon, so just acknowledging what's happening in some nice flavorful way is good enough i think 20160915 22:09:05< celmin> I'm thinking that replacing the first word with something else might fix the awkwardness. Assault from the air? Striking from the air? I dunno. 20160915 22:09:44< tad_> It's is sorta hard to get. You have to NOT attack the gryphons at all at the mountain and kill the loyalist leader, while letting his troops attack them 20160915 22:10:35< tad_> "Assualt from the air? How delicious!" 20160915 22:10:47< celmin> It'd be cool to have achievements for things like this. 20160915 22:10:50< celmin> …what? 20160915 22:11:22< tad_> Li'sar: "Assault from the air? What a delicious idea!" 20160915 22:11:57< tad_> Then I can use her glad profile and it fits. 20160915 22:13:14< DeFender1031> tad_, i like. 20160915 22:13:22< celmin> Heh. 20160915 22:13:40< celmin> Yeah, that sounds good. Are we dropping the speculation about mounting them then? 20160915 22:14:04< tad_> yes 20160915 22:14:31< zookeeper> maybe "fascinating" instead? or "intriguing" 20160915 22:14:53< tad_> inruguing 20160915 22:16:26< DeFender1031> delightful? 20160915 22:17:55< DeFender1031> tad_, zookeeper, ^^^ 20160915 22:18:41-!- mjs-de [~mjs-de@x4e31b778.dyn.telefonica.de] has joined #wesnoth-dev 20160915 22:18:43< tad_> OK. merged and pushed. I like it 20160915 22:20:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [] 20160915 22:22:07-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 22:23:18< tad_> celmin: No, about special locations and id= .. should I pull my new tag? does SLF have an id= so we can read map special locations into WML? If so, I can yank that commit on PR 775 and think it's ready for gfgtdf to merge if he's happy 20160915 22:23:32< celmin> I'll check in a bit. 20160915 22:23:39< celmin> Most likely after supper. 20160915 22:24:41< tad_> I should ponder dinner as well. Wife pulled an all-nighter sewing and is napping but she's gonna start grumbling in a bit ... 20160915 22:27:34-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20160915 22:31:28-!- mjs-de [~mjs-de@x4e31b778.dyn.telefonica.de] has quit [Remote host closed the connection] 20160915 22:31:57-!- Appleman1234_ is now known as Appleman1234 20160915 22:37:05-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds] 20160915 22:40:03-!- atarocch [~atarocch@HSI-KBW-46-223-95-173.hsi.kabel-badenwuerttemberg.de] has quit [Remote host closed the connection] 20160915 22:41:34 * tad_ was thinking about SLF and id= .. location_id= might be needed if SLF and SUF could be confused and id= already means a unit ID for some tags. [scroll_to] id= for example .. and we' want to be sure store_location gives back x,y so we can $var it into a [unit]x,y= or [move_unit] to_x,to_y= 20160915 22:44:25< zookeeper> there's no places where SUF and SLF are mixed in the same container, and i doubt there can be any ambiguity elsewhere either if an id= was added to SLF, but i guess it feels a bit... potentially confusing, somehow 20160915 22:46:24< gfgtdf> i have no strong opinion on whether it shoudl be called id= or location_id= 20160915 22:46:42< tad_> 'potentially confusing' --> how everything is designed :J 20160915 22:47:57< tad_> Like, I'm going to name the locations by the unit name so [unit]id=tad name=Tad placement=tad and figure THAT out! 20160915 22:48:50< zookeeper> oh no 20160915 22:49:36< zookeeper> ok so are special locations intended to be limited to a single hex? they can't refer to areas? 20160915 22:49:57< zookeeper> i'm guessing single hexes, but... 20160915 22:49:59-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev 20160915 22:50:29< tad_> That's how they came to me. A side starting location is [side] side=1 id=delfador name=Delfador placement=1 20160915 22:50:29< shadowm> SLF can filter on time areas and time areas don't need to define a custom ToD schedule, btw. 20160915 22:50:34< zookeeper> and it's been a long time since i've seen anyone else use the :J smiley 20160915 22:51:58< tad_> The thing is, for [side]side=1 .. placement=1 is implied by the engine 20160915 22:52:14< gfgtdf> shadowm: you sure they dont? the add_time_area which calls time_of_day::parse_times looks liek it alwas generated adummy scneario for the area 20160915 22:52:38< shadowm> "Dummy scenario". What? 20160915 22:52:53< gfgtdf> dummy time scedule i meant 20160915 22:53:40< shadowm> I'd have to recheck but I have used this in my campaign and I don't remember seeing anything weird. 20160915 22:53:46< tad_> gfgtdf: time areas are loaded before [side] .. hmm .. .oO(Don't get there, you know ZooKeeper is gonna deny the change.) 20160915 22:54:03< gfgtdf> shadowm: maybe you explicitly specified the defult time scedule in that aera ? 20160915 22:54:18< shadowm> I'd have to recheck. 20160915 22:54:21< shadowm> This was also intended by fabi to be used like this to assign ids to regions in the map editor. 20160915 22:55:28< mattsc> tad_, celmin: thanks for the comments.celmin: I replied to your question in another comment. 20160915 22:56:48< zookeeper> since special locations are, in the map data, written as prefixes of individual hex terrain codes (just like starting positions), i guess it makes sense that they're inherently limited to single hexes only and you can't paint multi-hex areas with them 20160915 22:58:23 * zookeeper shrugs 20160915 22:58:38< gfgtdf> shadowm: yes fabi said something liek that but it didnt wokr as expected when i tested it in the editor. 20160915 22:59:01< gfgtdf> zookeeper: yesspecial locations are basically a generiasation of startign locations 20160915 22:59:03< zookeeper> tad_, what's this about placement= though? can you use SingleUnitWML placement=nameofspeciallocation already or is that just planned? 20160915 22:59:18< tad_> IT works now. 20160915 22:59:22< zookeeper> right 20160915 22:59:27< zookeeper> someone didn't document it then 20160915 22:59:29< tad_> Well, sorta 20160915 23:00:12< tad_> Well, it worked, was wrong, got broken and someone wrapped it in 'if (false)' and I'm trying to fix that (on the map side .. [unit] side works fine) 20160915 23:07:29< gfgtdf> tad_: what what wrapped in if (false) was not really related to placment= handling in [unit] 20160915 23:08:49< zookeeper> mmkay 20160915 23:09:04< zookeeper> well, i'm out, will look at the PR's and stuff again tomorrow... 20160915 23:10:18< celmin> I think maybe location sets and time areas should be decoupled. 20160915 23:10:33< celmin> So that you define the area with [area] and then only reference it in [time_area]. 20160915 23:10:36< tad_> gfgtdf: It disabled storing the map special locations which placement= can use if they're there. Work fine in the main map. This is for [terrain_mask] 20160915 23:10:43< celmin> (Though I guess you could still do both at once in [time_area].) 20160915 23:13:29-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 250 seconds] 20160915 23:16:54-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed] 20160915 23:20:06-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev 20160915 23:30:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 23:35:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 23:38:10< celmin> tad_, gfgtdf: I don't see any support for special locations in SLF. https://github.com/wesnoth/wesnoth/blob/master/src/terrain/filter.cpp#L102 20160915 23:38:37< celmin> Ah wait. 20160915 23:38:52< gfgtdf> celmin: line 155 20160915 23:38:59< celmin> I missed it because of an unexpected indentation level: https://github.com/wesnoth/wesnoth/blob/master/src/terrain/filter.cpp#L155 20160915 23:39:21< celmin> What's ignore_xy? 20160915 23:40:13< gfgtdf> celmin: when match_internal is called by get_locations then match_internal shoudlnt filer things that get_locatiosn already did 20160915 23:40:44< celmin> Huh? 20160915 23:41:09< celmin> So basically part of the logic is duplicated as an optimization? 20160915 23:42:24< gfgtdf> celmin: ye 20160915 23:42:46< tad_> So I should test that [scroll_to] honors location_id= and [store_location] gives back x,y in the variable and if both are good I will pull the new tag? 20160915 23:43:12< celmin> I guess that makes sense though. No point looping over the whole map when exact coordinates were specified. 20160915 23:43:28< celmin> tad_: Sounds good to me. 20160915 23:44:40< celmin> gfgtdf: Do you know how to detect (in Lua or otherwise) whether a given map location is visible onscreen? 20160915 23:45:22< tad_> [scroll_to] honors location_id= .. 20160915 23:45:32< gfgtdf> celmin: hmm no i dont 20160915 23:45:44< gfgtdf> celmin: mostlikeley there is not function for that yet 20160915 23:46:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection] 20160915 23:46:12< celmin> Maybe I could look back at the C++ [message] implementation... 20160915 23:48:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev 20160915 23:49:21< tad_> OK. Give me a couple minutes to clear it out and push PR 775 without the new tag. 20160915 23:51:53< tad_> gfgtdf: PR 775 should be ready to merge once you're happy 20160915 23:56:09-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc] --- Log closed Fri Sep 16 00:00:36 2016