--- Log opened Thu May 04 00:00:04 2017
20170504 00:05:06< irker576> wesnoth: Charles Dang wesnoth:master e5b09de2c606 / src/gui/dialogs/multiplayer/mp_alerts_options.cpp: MP Alerts Options: simplify an overload disambiguation https://github.com/wesnoth/wesnoth/commit/e5b09de2c60627f9ec0ea543877b9ed6090c81d4
20170504 00:05:09< irker576> wesnoth: Charles Dang wesnoth:master e4f03fe4599c / / (135 files in 26 dirs): Moved all preferences source files into a single folder https://github.com/wesnoth/wesnoth/commit/e4f03fe4599c6969a8881955805fbfcaed3b01b3
20170504 00:33:55< vultraz_iOS> celticminstrel: seems it is possible tohave a memory leak even with a shared ptr
20170504 00:36:33< vultraz_iOS> problematic line seems to be current_track_list.emplace_back(new music_track(track));...
20170504 00:36:42< vultraz_iOS> track is a function-local object
20170504 00:37:10< vultraz_iOS> utilizing the copy ctor I guess
20170504 00:37:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 00:38:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 00:40:23< vultraz_iOS> wait
20170504 00:40:39< vultraz_iOS> oh
20170504 00:43:19< vultraz_iOS> yeah im not sure what's going on here..
20170504 00:44:07< vultraz_iOS> maybe the tracks aren't deleted properly?
20170504 00:50:31< celticminstrel> But they're managed by a shared_ptr. How would they not be deleted?
20170504 00:50:43< vultraz_iOS> i dunno
20170504 00:50:57< vultraz_iOS> it's possible to leak if something keeps it alive
20170504 00:51:05< vultraz_iOS> but i don't see what
20170504 00:54:38< celticminstrel> The only thing that could keep it alive would be another shared_ptr, which would then delete it, unless shared_ptr has something akin to unique_ptr::release
20170504 00:55:08< vultraz_iOS> but there isn't another shared ptr
20170504 00:55:17< celticminstrel> Right.
20170504 01:21:26-!- gfgtdf_ [~chatzilla@x4e36a3c9.dyn.telefonica.de] has joined #wesnoth-dev
20170504 01:24:25-!- gfgtdf [~chatzilla@x50abe4c3.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170504 01:24:36-!- gfgtdf_ is now known as gfgtdf
20170504 01:33:23< gfgtdf> celticminstrel: my current plan is to fix the appveyor build before i finish #1010. But i probably cannot finish the appveyor buld since for some reason boost fails to build here
20170504 01:36:43-!- gfgtdf [~chatzilla@x4e36a3c9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 53.0/20170413192749]]
20170504 01:54:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170504 02:02:59< vultraz_iOS> how do the tests handle assert?
20170504 02:03:58< celticminstrel> ???
20170504 02:04:17< vultraz_iOS> trying to figure out what's happening with https://github.com/wesnoth/wesnoth/issues/1013
20170504 02:05:21< celticminstrel> Some function returned the address of a local variable, it seems.
20170504 02:06:15< celticminstrel> Or it could be something like "T* p = nullptr; if(some_condition) {T thing; p = &thing;}"
20170504 02:06:17< vultraz_iOS> oh, I see, I stored the list as a reference so i guess the object it's pointing to gets destroyed?
20170504 02:06:42< celticminstrel> What line?
20170504 02:06:55< vultraz_iOS> tests 991
20170504 02:07:04< celticminstrel> ...
20170504 02:07:18< vultraz_iOS> what?
20170504 02:07:30< celticminstrel> Okay, line 109 apparently.
20170504 02:07:42< celticminstrel> According to the log on the issue.
20170504 02:07:56< celticminstrel> Oh.
20170504 02:08:08< celticminstrel> So that's only where it's detected, not the root cause.
20170504 02:08:21< vultraz_iOS> i changed map_generators_ to be a reference in the dialog
20170504 02:08:32< vultraz_iOS> so im guessing since the tests pass in a local variable it gets destroyed
20170504 02:09:00< vultraz_iOS> that is, map_generators should move out of the create() function?
20170504 02:09:27< celticminstrel> Ah, yes. Make it a member variable instead.
20170504 02:11:40< irker576> wesnoth: Charles Dang wesnoth:master ce11f4855e89 / src/tests/gui/test_gui2.cpp: Attempt to fix tests (fixup 7d9b3f44ad8) https://github.com/wesnoth/wesnoth/commit/ce11f4855e897e49550ac4ac6fb185dc652eae48
20170504 02:41:55< irker576> wesnoth: Celtic Minstrel wesnoth:lua_require 08b381d4e424 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/08b381d4e4242ca916fcc71922d42ffc81933803
20170504 02:43:14< celticminstrel> vultraz_iOS: Feel free to merge if you think it's okay. ^
20170504 02:48:11< Aginor> hey guys
20170504 03:01:58< Aginor> I'd like to resign my committership
20170504 03:02:15< celticminstrel> ...!?
20170504 03:02:47< celticminstrel> Does that mean you've decided not to contribute anymore?
20170504 03:04:05-!- sevu [~Unknown@p5485651C.dip0.t-ipconnect.de] has quit [Disconnected by services]
20170504 03:04:05-!- shiki [~Unknown@p54857127.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170504 03:04:21-!- vultraz_iOS_ [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170504 03:05:42< celticminstrel> vultraz_iOS_: Feel free to merge PR 1021 if you think it's okay.
20170504 03:05:51< Aginor> I'm already not contributing in any meaningful way, so yes
20170504 03:06:41< celticminstrel> :(
20170504 03:09:09< Aginor> I shall go and remove myself from the github organisation and send an email to the mailing list, but please revoke my other accesses or forum status as is appropriate
20170504 03:10:59< Aginor> I would like to thank all of you though, it's been fun to hack on stuff together
20170504 03:11:54-!- Netsplit *.net <-> *.split quits: Appleman1234, vultraz_iOS, bumbadadabum
20170504 03:12:09-!- vultraz_iOS_ is now known as vultraz_iOS
20170504 03:19:16-!- Netsplit over, joins: bumbadadabum
20170504 03:23:20< Aginor> I've now left wesnoth on github
20170504 03:26:02-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has joined #wesnoth-dev
20170504 03:26:02-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has quit [Changing host]
20170504 03:26:02-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has joined #wesnoth-dev
20170504 03:26:27< Aginor> I've thought a lot about it over the last 6 months and I think it's better this way
20170504 03:26:58< Aginor> I may however contribute the occasional odd PR
20170504 03:35:35< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 3215451f0d65 / src/credentials.cpp: Escape/unescape only when saving/loading https://github.com/wesnoth/wesnoth/commit/3215451f0d6574561e71bf754d6d1236d76d785c
20170504 03:35:37< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt b7d303ccadf8 / src/credentials.cpp: Use RC4 instead of AES https://github.com/wesnoth/wesnoth/commit/b7d303ccadf8e9941febd001af5344be43681d43
20170504 03:42:54-!- Appleman1234 [~Appleman1@pl23964.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170504 03:44:35< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 328afe4a6730 / src/credentials.cpp: Reduce reinterpret_casts by using vector<unsigned char> https://github.com/wesnoth/wesnoth/commit/328afe4a67309ebf6ba8645da924ed14d1f31bd0
20170504 03:44:45< celticminstrel> I see.
20170504 03:44:53 * celticminstrel pokes vultraz_iOS.
20170504 04:05:33-!- JyrkiVesterinen [~JyrkiVest@87-100-215-124.bb.dnainternet.fi] has joined #wesnoth-dev
20170504 04:37:28< vultraz_iOS> hm?
20170504 04:43:45< vultraz_iOS> celticminstrel: does a virtual dtor need to have anything in it - ie virtual ~() {} or can it just be declared: virtual ~T();
20170504 04:44:14< celticminstrel> It needs to be defined.
20170504 04:44:19< JyrkiVesterinen> A virtual destructor needs a definition like any other destructor.
20170504 04:44:25< celticminstrel> Unless it's pure virtual, of course.
20170504 04:44:37< celticminstrel> Or wait, do those need to be defined too?
20170504 04:44:52< celticminstrel> I never make destructors pure virtual anyway, so I can't remember.
20170504 04:45:11< JyrkiVesterinen> "virtual ~T() {}" is the shortest option. Another possibility is "virtual ~T() = default;"
20170504 04:45:33< JyrkiVesterinen> AFAIK, pure virtual destructors don't need to be defined.
20170504 04:46:43< irker576> wesnoth: Charles Dang wesnoth:master ae85e4ca0239 / src/storyscreen/parser.hpp: Seems this should have a virtual dtor https://github.com/wesnoth/wesnoth/commit/ae85e4ca02392e10f6333412d130539955d3d53c
20170504 04:57:30< irker576> wesnoth: Severin Glöckner wesnoth:master 616f3b5d66cf / data/campaigns/tutorial/ (scenarios/01_Tutorial_part_1.cfg utils/utils.cfg): Take care about the player being unable to do anything if he https://github.com/wesnoth/wesnoth/commit/616f3b5d66cf1f8f6ab3a8edeedd03ed803e99c9
20170504 05:08:47< wedge009> gfgtdf: So it seems. 1 - Would be happy to do that, but is there no-one else near you willing to try this? You seem to be in an American TZ. 2 - I made an attempt at adding the remaining utility headers for 1.64 to VC15 branch. Don't know if that helps you.
20170504 05:10:39< celticminstrel> Was the MSVC project updated yet for the prefs reorganization?
20170504 05:13:06< JyrkiVesterinen> No, it wasn't.
20170504 05:16:33< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 5aba7691aa15 / src/credentials.cpp: Address feedback from @jyrkive https://github.com/wesnoth/wesnoth/commit/5aba7691aa155b81a930e65b09961dfff594df53
20170504 05:17:29< celticminstrel> Oh, you also asked to deploy more vectors.
20170504 05:17:37< JyrkiVesterinen> Indeed.
20170504 05:17:43< celticminstrel> Let's see if I can do that without uglifying everything...
20170504 05:18:55< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/commit/5aba7691aa155b81a930e65b09961dfff594df53#commitcomment-22011233
20170504 05:20:59< celticminstrel> JyrkiVesterinen: Is it okay if build_key takes strings and outputs a vector?
20170504 05:21:43< JyrkiVesterinen> That would be the best option.
20170504 05:21:53< JyrkiVesterinen> Its inputs are human-readable, but output is not.
20170504 05:22:05< JyrkiVesterinen> Hence, inputs should be strings, and output should be a vector.
20170504 05:31:03< celticminstrel> Hmm, wait, should decrypt return a string and encrypt take a string?
20170504 05:31:58< JyrkiVesterinen> Is it a part of decrypt()s API contract that it always returns human-readable text?
20170504 05:32:09< JyrkiVesterinen> IMO, it shouldn't be.
20170504 05:32:11< celticminstrel> I guess not.
20170504 05:32:17< JyrkiVesterinen> In which case it should return a vector.
20170504 05:32:33< celticminstrel> In practice it'll mostly be human-readable, but there's no reason to assume it has to be.
20170504 05:35:23-!- Appleman1234 [~Appleman1@pl23964.ag1212.nttpc.ne.jp] has quit [Ping timeout: 255 seconds]
20170504 05:54:42< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 639a78df5ff3 / src/credentials.cpp: Address most recent feedback from @jyrkive https://github.com/wesnoth/wesnoth/commit/639a78df5ff3b884458f73a77e5fa52456862d5f
20170504 05:54:44< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 779d9d4bc656 / src/credentials.cpp: Use vectors whenever handling keys or possibly-encrypted data https://github.com/wesnoth/wesnoth/commit/779d9d4bc6566254f052fbe9c9f425434615feb2
20170504 05:56:49< irker576> wesnoth: Wedge009 wesnoth:master 265a0e9bd51b / projectfiles/VC12/ (wesnoth.vcxproj wesnoth.vcxproj.filters): Updating VC project files for e4f03fe4. https://github.com/wesnoth/wesnoth/commit/265a0e9bd51b871ab47702f09e8405d3c104f64b
20170504 05:57:54< celticminstrel> Alright, now any idea how to get the PR990 CMake build to work?
20170504 05:58:56< celticminstrel> The Mac build is also troublesome... because for some strange reason the headers aren't accessible even though I explicitly installed it and added the paths to the environment...
20170504 05:59:32< celticminstrel> ISTR that scons does honour CFLAGS and LDFLAGS at least?
20170504 05:59:49< JyrkiVesterinen> Regarding CMake, if you can't fix it before then, I can investigate the problem next week.
20170504 05:59:56< celticminstrel> Okay.
20170504 06:00:12< celticminstrel> I'll try to focus on the Mac issue then...
20170504 06:00:23< celticminstrel> I probably just did it wrong in CMake.
20170504 06:02:48< celticminstrel> vultraz_iOS (or anyone else I guess): Any objection to merging 1021?
20170504 06:02:48-!- mjs-de [~mjs-de@x4db5f238.dyn.telefonica.de] has joined #wesnoth-dev
20170504 06:03:16< vultraz_iOS> as long as it works
20170504 06:03:34< celticminstrel> It seemed to work when I tested it.
20170504 06:03:46< celticminstrel> And this is something that would have major consequences if it failed, so...
20170504 06:04:07< celticminstrel> It'd break literally all Lua if it didn't work.
20170504 06:07:37< celticminstrel> vultraz_iOS: And what about 987 / 984?
20170504 06:08:18< vultraz_iOS> sure
20170504 06:08:42< celticminstrel> I guess we should check zookeeper's approval for 983.
20170504 06:11:00< irker576> wesnoth: Wedge009 wesnoth:master 08420e6d1044 / data/core/units/undead/Spirit_Shadow.cfg: Invert alpha animation for Shadow (bug #14503) (#984) https://github.com/wesnoth/wesnoth/commit/08420e6d1044c2b036e267cb2ad4306a66a94e48
20170504 06:11:14< irker576> wesnoth: Wedge009 wesnoth:master 713995276b5d / src/units/drawer.cpp: Reduce invisible units' transparency (#987) https://github.com/wesnoth/wesnoth/commit/713995276b5d08fa999da12c61f8df8296d79c36
20170504 06:11:27< irker576> wesnoth: Celtic Minstrel wesnoth:master 37c0ccafa3be / src/scripting/lua_kernel_base.cpp: Allow wesnoth.dofile to forward arbitrary arguments to the file's "..." https://github.com/wesnoth/wesnoth/commit/37c0ccafa3bec08592a78b6fc1b77d25e45311f8
20170504 06:11:29< irker576> wesnoth: Celtic Minstrel wesnoth:master 072ddc4dc121 / data/lua/package.lua src/scripting/lua_kernel_base.cpp: Reimplement wesnoth.require function in Lua https://github.com/wesnoth/wesnoth/commit/072ddc4dc121f7bb0b264f1c8f4d1c506b62dd90
20170504 06:11:31< irker576> wesnoth: Celtic Minstrel wesnoth:master 6b74f91b2feb / data/lua/package.lua: Allow shorter module name specification in wesnoth.require https://github.com/wesnoth/wesnoth/commit/6b74f91b2feb4bca38f8675d141f41a455525e17
20170504 06:11:33< irker576> wesnoth: Celtic Minstrel wesnoth:master 8d4cf3cf62f0 / data/ (98 files in 21 dirs): Shorten requires where possible https://github.com/wesnoth/wesnoth/commit/8d4cf3cf62f0e402cb8356dcd8a8155daf3e5bd8
20170504 06:11:35< irker576> wesnoth: Celtic Minstrel wesnoth:master 448dd5d0b494 / src/scripting/lua_fileops.cpp: Minor reformatting and comment fixup https://github.com/wesnoth/wesnoth/commit/448dd5d0b4944f48c92e5ac815cb7db2754e2409
20170504 06:11:37< irker576> wesnoth: Celtic Minstrel wesnoth:master e2a2790182c1 / src/scripting/lua_fileops.cpp: Add flag to wesnoth.have_file to require it to be a real file, not a directory https://github.com/wesnoth/wesnoth/commit/e2a2790182c12e539b4976713777b2d908ff5b87
20170504 06:11:39< irker576> wesnoth: Celtic Minstrel wesnoth:master 3eae385f3f18 / src/scripting/lua_fileops.cpp: Allow wesnoth.read_file to list a directory contents https://github.com/wesnoth/wesnoth/commit/3eae385f3f1848ce3b07e21e35a3b0bc887c2dda
20170504 06:11:41< irker576> wesnoth: Celtic Minstrel wesnoth:master 690acf9b790e / data/lua/ (package.lua wml-tags.lua): Allow wesnoth.require to load a whole directory in one call https://github.com/wesnoth/wesnoth/commit/690acf9b790e185b119f0a8ed64ead7cac496550
20170504 06:11:43< irker576> wesnoth: Celtic Minstrel wesnoth:master e1233fd0f244 / data/lua/ (9 files in 2 dirs): Split several of the larger WML tags into their own file https://github.com/wesnoth/wesnoth/commit/e1233fd0f244c12727bb9be349ee12dd6dac2edc
20170504 06:11:45< irker576> wesnoth: Celtic Minstrel wesnoth:master 5da2d71a205b / data/ (ai/micro_ais/micro_ai_wml_tag.lua core/_main.cfg lua/wml/micro_ai.lua): Move [micro_ai] implementation to data/lua/wml https://github.com/wesnoth/wesnoth/commit/5da2d71a205b6e2411972e403fe66733fb653de0
20170504 06:11:47< irker576> wesnoth: Celtic Minstrel wesnoth:master badc2d56d0cd / data/lua/wml/micro_ai.lua: Load all MicroAI definitions in a single line https://github.com/wesnoth/wesnoth/commit/badc2d56d0cd61a16743a1373b72d6563cb2a7a2
20170504 06:11:49< irker576> wesnoth: Celtic Minstrel wesnoth:master 9ad5a5698437 / data/lua/wml/test_condition.lua: Fix lua error in [test_condition] https://github.com/wesnoth/wesnoth/commit/9ad5a56984374083065cbbba4d6af0cea1c845cb
20170504 06:11:51< irker576> wesnoth: Celtic Minstrel wesnoth:master 08b381d4e424 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/08b381d4e4242ca916fcc71922d42ffc81933803
20170504 06:11:53< irker576> wesnoth: Celtic Minstrel wesnoth:master 787a46820ee7 / / (112 files in 24 dirs): Merge pull request #1021 from wesnoth/lua_require https://github.com/wesnoth/wesnoth/commit/787a46820ee71e5538cbcd1b431c6560e4d94b47
20170504 06:17:03< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt b93dc472031f / INSTALL.md: Update INSTALL https://github.com/wesnoth/wesnoth/commit/b93dc472031fb0248f7207347e51e10bc8cb23bf
20170504 06:18:40-!- mjs-de [~mjs-de@x4db5f238.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170504 06:23:08-!- JyrkiVesterinen [~JyrkiVest@87-100-215-124.bb.dnainternet.fi] has quit [Quit: .]
20170504 06:28:12< celticminstrel> Huh, apparently you can edit the base branch of a PR. I didn't know that.
20170504 06:28:39< celticminstrel> That is, the branch it'll be merged into.
20170504 06:28:46-!- celticminstrel is now known as celmin|sleep
20170504 06:44:21-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170504 06:47:25-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170504 06:48:22-!- Appleman1234 [~Appleman1@pl43828.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170504 06:55:12-!- JyrkiVesterinen [~JyrkiVest@85-76-67-254-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170504 07:17:54< zookeeper> #983 is fine
20170504 07:20:00< vultraz_iOS> celmin|sleep: master passed again before your merged
20170504 07:20:02< vultraz_iOS> merges
20170504 07:21:19< zookeeper> Aginor, that's unfortunate, but if you're sure...
20170504 07:23:14< zookeeper> i don't think we've really ever modified access permissions or forum groups and such when they've retired, though (except for moderators etc).
20170504 07:23:27< zookeeper> s/they/people
20170504 07:31:29< Aginor> zookeeper: I am :/
20170504 07:32:07< Aginor> I've already left the GH organisation so I don't have commit access any longer
20170504 07:32:33< wedge009> Hope to see you around again, Aginor.
20170504 07:33:21< vultraz_iOS> Aginor: well, we must thank you for elevating us to SDL2.
20170504 07:33:22< Aginor> I'll loiter around here a bit longer in case anyone has any questions for me
20170504 07:33:34< vultraz_iOS> I doubt it would have happened without your efforts
20170504 07:33:43 * wedge009 concurs!
20170504 07:34:29< zookeeper> Aginor, great! how about sdl::draw_rectangle(clip.x, clip.y, clip.w, clip.h, 0x00000000, screen); between these two lines? https://github.com/wesnoth/wesnoth/blob/master/src/display.cpp#L1496-L1497
20170504 07:34:45< zookeeper> seems to fix the background problem i was having, but i dunno if there's something wrong with it.
20170504 07:35:57< Aginor> draw_background(surface screen,
20170504 07:36:06< Aginor> you need that to be a reference or it'll crash
20170504 07:36:22< Aginor> always pass surfaces as references, never by value
20170504 07:37:11< zookeeper> okay, but... that's not a problem for sdl_blit?
20170504 07:37:29< Aginor> resize the screen, and it'll crash in that function
20170504 07:37:33< vultraz_iOS> are you sure you don't just want to fill the screen with black first
20170504 07:37:44< zookeeper> i'm not sure
20170504 07:38:13< JyrkiVesterinen> To me, it sounds like just filling the entire screen with black would be best.
20170504 07:38:20< vultraz_iOS> in which case you want CVideo::get_singleton().get_window()->fill(0,0,0,255);
20170504 07:38:21< JyrkiVesterinen> That's how many game engines operate.
20170504 07:38:26< vultraz_iOS> I just noticed something, though..
20170504 07:38:36< vultraz_iOS> in the window class the default alpha argument for fill is 0...
20170504 07:38:56< Aginor> I too advocate filling with black
20170504 07:39:03< vultraz_iOS> zookeeper: I wonder if changing sdl/window.cpp:56 to `fill(0,0,0,255)` will fix your problems
20170504 07:39:07< zookeeper> i've interpreted those loops as just filling the background area, in which case filling the entire screen might fill over something else that was already drawn prior
20170504 07:39:07< vultraz_iOS> zookeeper: please try
20170504 07:39:37< Aginor> or draw a rectangle over the area you intend to redraw over
20170504 07:39:37< zookeeper> will do
20170504 07:40:17< JyrkiVesterinen> zookeeper: Yeah, we have a weird rendering pipeline that likes to depend on the previous content of the framebuffer...
20170504 07:40:50< Aginor> JyrkiVesterinen: yes, I have spent some effort into trying to straightening it out, but didn't get terribly far
20170504 07:41:06< Aginor> I had a promising branch, but it got too hard to maintain
20170504 07:41:24< Aginor> JyrkiVesterinen: that's unfortunately a hard blocker for gettining meaningful HW accell
20170504 07:41:59< JyrkiVesterinen> How so? OpenGL doesn't force you to clear the framebuffer when you start drawing a new frame.
20170504 07:42:11< Aginor> zookeeper: https://wiki.libsdl.org/SDL_BlendMode
20170504 07:42:31< Aginor> JyrkiVesterinen: it copies the screen into new surfaces, draws soemthing, then copies it back
20170504 07:42:41< Aginor> so it can't all live in video ram
20170504 07:42:50< Aginor> which really kills performance
20170504 07:43:03< JyrkiVesterinen> Yes, the unrender stuff. I'm aware of that.
20170504 07:43:23-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has quit [Ping timeout: 260 seconds]
20170504 07:43:46< JyrkiVesterinen> I thought you were talking about how the game doesn't redraw parts that haven't changed. It's abnormal, but not a blocker for HW acceleration.
20170504 07:43:53< Aginor> no
20170504 07:44:07< Aginor> I think that's an attempt at performance optimisation
20170504 07:46:48< zookeeper> well, isn't that quite a bit more than an attempt when software rendering is used?
20170504 07:47:34< zookeeper> surely performance would be a lot worse if the game just software-rendered the whole screen every frame? (even if it would make hardware-rendering much easier to switch to)
20170504 07:48:59< Aginor> zookeeper: depending on the map and what's on it, it actually does that multiple times per frame in an attempt to avoid a full redraw
20170504 07:52:33< Aginor> anyway
20170504 07:52:45< Aginor> thanks for the opportunity to contribute to wesnoth
20170504 07:53:15< wedge009> Always welcome. ;)
20170504 07:54:16< irker576> wesnoth: Charles Dang wesnoth:master 8eec51ca69c0 / src/gui/core/event/ (dispatcher.cpp dispatcher.hpp): GUI2: drop t prefix from hotkey function typedef https://github.com/wesnoth/wesnoth/commit/8eec51ca69c0101bfd14ddaabcc0693f1900873c
20170504 07:54:19< irker576> wesnoth: Charles Dang wesnoth:master a967e4bb4623 / src/display.cpp: Removed some evil commented out code https://github.com/wesnoth/wesnoth/commit/a967e4bb46234b329431fd89ad88ac7a402adac2
20170504 07:55:22< JyrkiVesterinen> Hey. The code you removed can potentially be useful for reproducing bugs which only occur on slow PCs.
20170504 07:55:25< zookeeper> vultraz_iOS, your suggestion seems to fix the problem too
20170504 07:55:49< zookeeper> JyrkiVesterinen, but it's evil because it makes things slower?
20170504 07:56:01< vultraz_iOS> zookeeper: sweet :D
20170504 07:56:14< JyrkiVesterinen> Only if someone intentionally removes the commenting...
20170504 07:56:33< vultraz_iOS> zookeeper: if you commit that, please change the default value for the alpha argument in fill to 255
20170504 07:56:44< vultraz_iOS> though don't use 255, use the SDL_ALPHA_OPAQUE macro
20170504 07:57:03< vultraz_iOS> similarly in the call site if you call it explicitly 
20170504 07:59:42< vultraz_iOS> (I'd say use our own ALPHA_OPAQUE macro but I don't want to pull in color.hpp as a dependency)
20170504 08:00:32< zookeeper> eh, actually the bug doesn't reproduce now in the first place, so that's why i thought it worked...
20170504 08:01:01< zookeeper> what is this sorcery
20170504 08:02:35< mattiaskrgr> remember: magic bugs have a 70% base chance of being undetected 
20170504 08:03:01< zookeeper> indeed
20170504 08:04:29-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has quit [Quit: nurupo]
20170504 08:05:00-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20170504 08:05:36< vultraz_iOS> Aginor: still around for a sec?
20170504 08:06:06< Aginor> indeed
20170504 08:08:36< vultraz_iOS> Aginor: your recent hotkey change broke the lua console hotkey in the titlescreen. if I invert the condition for is_uncomposable (ie, !CKey::is_uncomposable) here: https://github.com/wesnoth/wesnoth/blob/master/src/hotkey/hotkey_item.cpp#L294 it works again, but I'm not sure if that's the right thing to do
20170504 08:09:30< Aginor> that'll really break stuff
20170504 08:10:09< Aginor> can it be bound on its own?
20170504 08:10:28< Aginor> yes - problem with the saving path. no - problem witht he creation path
20170504 08:11:03< vultraz_iOS> uhhh....
20170504 08:12:45< Aginor> inverting that condition will break a lot of hotkeys
20170504 08:12:57< zookeeper> well, as said, the background bug is gone now, so unless it's just some glitch with my current executable i guess it's fine as-is -.-
20170504 08:13:50< vultraz_iOS> adding a new hotkey for it doesn't work either
20170504 08:14:16< mattiaskrgr> mmh sometimes there's a strange bug were you can't click your units
20170504 08:14:18< mattiaskrgr> is this known? 
20170504 08:14:25< vultraz_iOS> o-O
20170504 08:14:40< mattiaskrgr> oh I unfocussed and refocussed the game window now it works 
20170504 08:14:41< mattiaskrgr> fun 
20170504 08:14:50< JyrkiVesterinen> mattiaskrgr: I recall there has been reports about such a bug.
20170504 08:14:57< mattiaskrgr> mmh
20170504 08:15:04< JyrkiVesterinen> As it's very rare, it's almost impossible to investigate.
20170504 08:15:08< mattiaskrgr> :( 
20170504 08:15:57< vultraz_iOS> it has to do with conflicting event contexts
20170504 08:16:49< vultraz_iOS> if the foremost event context isn't the one handling the gamemap you can't select anything
20170504 08:17:44< vultraz_iOS> Aginor: the default lua console hotkey is "`" if that makes any difference
20170504 08:18:02< vultraz_iOS> other hotkeys like F5 work in the titlescreen
20170504 08:18:10< zookeeper> ohh wait dumb me! the background bug was seeemingly gone only because i forgot my local images are 24-bit which is another workaround for it...
20170504 08:18:20< zookeeper> back to testing, then...
20170504 08:18:23< vultraz_iOS> heh
20170504 08:18:40< Aginor> vultraz_iOS: you could add the appropriate key to the list of uncomposable keys
20170504 08:19:29< Aginor> I don't think anyone is likely to want ẽ as a hotkey
20170504 08:19:36< Aginor> hmm, wrong composition
20170504 08:19:52< Aginor> but you get the idèa
20170504 08:20:34< zookeeper> vultraz_iOS, nope, your suggestion doesn't fix the background bug
20170504 08:20:37< vultraz_iOS> fuck
20170504 08:20:45< zookeeper> (i think i actually already tried that before)
20170504 08:21:11< Aginor> vultraz_iOS: SDLK_BACKQUOTE
20170504 08:21:17< Aginor> https://wiki.libsdl.org/SDL_Keycode
20170504 08:21:19< vultraz_iOS> already got it, thanks
20170504 08:21:33< Aginor> I may have missed others
20170504 08:21:42< Aginor> but I hope not
20170504 08:22:37< vultraz_iOS> now to figure out why a hotkey is saved twice in prefs and once without the key= code
20170504 08:23:45< vultraz_iOS> oh wait
20170504 08:23:51< vultraz_iOS> the first is the default
20170504 08:23:58< vultraz_iOS> the second is the modified value
20170504 08:24:08< vultraz_iOS> but without key=
20170504 08:26:21< zookeeper> Aginor, vultraz_iOS, btw, i don't get any crashes when resizing the window with my suggested background fix
20170504 08:26:41< vultraz_iOS> i still recommend filling the window
20170504 08:26:51< Aginor> zookeeper: it's racy, make it a reference
20170504 08:27:45< Aginor> in the method signature
20170504 08:27:55< vultraz_iOS> Aginor: for some reason, keycode_ is SDLK_UNKNOWN in save_helper :/ that;s why hotkeys aren't persisting
20170504 08:29:37< Aginor> make it save text_
20170504 08:31:37< vultraz_iOS> 	if(!text_.empty()) {
20170504 08:31:37< vultraz_iOS> 		item["key"] = text_;
20170504 08:31:37< vultraz_iOS> 	}
20170504 08:31:39< vultraz_iOS> you mean?
20170504 08:31:43< vultraz_iOS> (that seems to work)
20170504 08:31:47< Aginor> yes
20170504 08:31:51< zookeeper> Aginor, turns out i don't know how to make it a reference
20170504 08:32:06< Aginor> zookeeper: surface &surf
20170504 08:32:20< Aginor> or surface &screen in this instance
20170504 08:32:27< vultraz_iOS> Aginor: thanks for helping :)
20170504 08:33:27< zookeeper> Aginor, sure, but that's not enough
20170504 08:35:04< zookeeper> ok, i think i got it... just had to remove a const somewhere else
20170504 08:38:46-!- shiki [~Unknown@p54857127.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20170504 08:42:34< zookeeper> eh. but now i can't get my fix to work at all, reference or not.
20170504 08:42:37 * zookeeper dislikes
20170504 08:43:00< Aginor> the reference is orthogonal to your bug ;)
20170504 08:46:56-!- JyrkiVesterinen [~JyrkiVest@85-76-67-254-nat.elisa-mobile.fi] has quit [Quit: .]
20170504 09:12:49-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170504 09:17:30-!- JyrkiVesterinen [~JyrkiVest@85-76-67-254-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170504 09:35:05< irker576> wesnoth: Charles Dang wesnoth:master 14068bda5696 / src/key.cpp: Add SDLK_BACKQUOTE to the list of uncomposable hotkeys https://github.com/wesnoth/wesnoth/commit/14068bda56965630b4afbd0631f21afdfbb3007f
20170504 09:35:08< irker576> wesnoth: Charles Dang wesnoth:master 3ba0d41d3793 / src/hotkey/hotkey_item.cpp: Fix hotkeys not persisting between wesnoth sessions (fixes #1016) https://github.com/wesnoth/wesnoth/commit/3ba0d41d379335fb1d7c8601b688a7f649374b57
20170504 09:38:39< irker576> wesnoth: sigurdfdragon wesnoth:master 329c29990294 / data/campaigns/An_Orcish_Incursion/scenarios/ (05_Linaera_the_Quick.cfg 06_A_Detour_through_the_Swamp.cfg): AOI: Inform the player that Linaera can recruit https://github.com/wesnoth/wesnoth/commit/329c29990294e82652374e7774442f5750b175b9
20170504 09:46:19-!- clavi [~clavi@v22017034422546657.goodsrv.de] has quit [Quit: ZNC - http://znc.in]
20170504 09:52:28-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20170504 09:52:39-!- clavi [~clavi@v22017034422546657.goodsrv.de] has joined #wesnoth-dev
20170504 10:10:56-!- Kwandulin [~Kwandulin@p200300760F6D80A5D1C63ED7267BA5CC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170504 10:41:02< zookeeper> ok, i got a (slightly different) fix that actually seems to work... i think i'll just push it and then you can mallet me if there's something wrong with it
20170504 10:42:09 * vultraz_iOS prepares squeaky hammer
20170504 10:43:28< irker576> wesnoth: ln-zookeeper wesnoth:master e89abe4b9610 / src/display.cpp: Fixed a subtle glitch by filling map background before blitting bitmaps https://github.com/wesnoth/wesnoth/commit/e89abe4b9610f6d1d6c199eb436a780d93d69b0f
20170504 10:44:12< vultraz_iOS> hmmm
20170504 10:44:17< vultraz_iOS> ok, perhaps this isn't needed
20170504 10:44:21< vultraz_iOS> *puts down hammer*
20170504 10:48:13-!- Alkenrinnstet [~alkenrinn@42.61.217.253] has joined #wesnoth-dev
20170504 11:15:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 11:21:49-!- Kwandulin [~Kwandulin@p200300760F6D80A5D1C63ED7267BA5CC.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170504 11:23:26< irker576> wesnoth: Charles Dang wesnoth:master dc0cd76b0eac / src/gui/widgets/ (window.cpp window.hpp): GUI2/Window: handle event distributor with a unique_ptr https://github.com/wesnoth/wesnoth/commit/dc0cd76b0eac392b1cdbd368fb57df37d60c7b11
20170504 11:23:28< irker576> wesnoth: Charles Dang wesnoth:master 039463e78cb0 / src/gui/core/event/ (distributor.cpp distributor.hpp): GUI2/Distributor: wrap ui_event types in a helper metastruct to reduce huge temp https://github.com/wesnoth/wesnoth/commit/039463e78cb0c3d4db9f9d1b8febeee9d1c022cd
20170504 11:55:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 11:56:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 12:16:39-!- travis-ci [~travis-ci@54.166.53.34] has joined #wesnoth-dev
20170504 12:16:40< travis-ci> wesnoth/wesnoth#13713 (master - a967e4b : Charles Dang): The build has errored.
20170504 12:16:40< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/228636624
20170504 12:16:40-!- travis-ci [~travis-ci@54.166.53.34] has left #wesnoth-dev []
20170504 12:21:04< JyrkiVesterinen> This is the error according to the log:
20170504 12:21:06< JyrkiVesterinen> 20170504 11:39:05 error scripting/lua: encountered a runtime error:
20170504 12:21:06< JyrkiVesterinen> [string "-- join.lua --..."]:19: attempt to index a function value (upvalue 'helper')
20170504 12:21:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 12:21:45 * zookeeper notes that vultraz_iOS never clarified why he removed the commented-out code
20170504 12:21:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 12:22:24< irker576> wesnoth: ln-zookeeper wesnoth:master 769a91c720e9 / data/campaigns/Son_Of_The_Black_Eye/maps/07_The_Desert_of_Death.map: SotBE S07: Added extra village to enemy camp, and updated mountains https://github.com/wesnoth/wesnoth/commit/769a91c720e9a483ba0aec2241c947920e13beb5
20170504 12:22:31< vultraz_iOS> zookeeper: code to make things slower is bad
20170504 12:23:02< JyrkiVesterinen> Even if it's commented out (and thus isn't run)?
20170504 12:23:31< zookeeper> JyrkiVesterinen, i guess you can just add it back? :p
20170504 12:24:11< JyrkiVesterinen> Locally? Yes, sure.
20170504 12:24:26< JyrkiVesterinen> In the repository? I'd rather not revert commits from other developers.
20170504 12:26:12< vultraz_iOS> I had figured if it was left there someone might think to uncomment it in the future.
20170504 12:26:54< zookeeper> vultraz_iOS, seriously, are you trolling or what?
20170504 12:27:04< JyrkiVesterinen> "Gee, the comment says that uncommenting this will simulate a slow PC. Maybe I should enable that by default."
20170504 12:27:20< JyrkiVesterinen> "I'm sure everyone will want to feel the slow PC experience!!"
20170504 12:27:57< zookeeper> yeah basically if your argument is "i'll remove this commented-out code that's clearly labeled as something no one with half a brain would uncomment if they don't know what they're doing, just so someone won't accidentally uncomment it and push that change" then... uh...
20170504 12:28:04< vultraz_iOS> Ok yeah when you out it that way it sounds absurd :p
20170504 12:29:38< vultraz_iOS> Put
20170504 12:30:20< vultraz_iOS> I guess I also figured since no one would ever want to use it one shouldn't leave it around
20170504 12:31:02< zookeeper> what makes you think no one would ever want to use it?
20170504 12:32:17< vultraz_iOS> Because I assumed no one would want to make things slower
20170504 12:32:56< JyrkiVesterinen> It may be useful to reproduce bugs which only occur on slow computers.
20170504 12:33:04< JyrkiVesterinen> Is it OK if I revert the commit?
20170504 12:33:26< zookeeper> vultraz_iOS, and at no point did you think that the fact the code is there is evidence that obviously someone has sometimes wanted to make things slower?
20170504 12:34:13< vultraz_iOS> Revert it if you want, I have no objections
20170504 12:34:45< vultraz_iOS> zookeeper: well, yes - but I figured it's not something one would want to do in the future nor something one should have been doing in the past
20170504 12:35:19< zookeeper> sounds a bit circular ;)
20170504 12:35:34< vultraz_iOS> Though now that I think about it what i assumed it was used for probably wasn't what it was used for
20170504 12:35:50< vultraz_iOS> Since I guess no one would probably have introduced code to cap performance
20170504 12:36:10< vultraz_iOS> But that's what I thought it was for
20170504 12:36:25< vultraz_iOS> In which case you can see why I would think one should not cap performance
20170504 12:36:59< irker576> wesnoth: Jyrki Vesterinen wesnoth:master 312743403b30 / src/display.cpp: Revert "Removed some evil commented out code" https://github.com/wesnoth/wesnoth/commit/312743403b3066536be5d306c81edb4571e640fe
20170504 12:39:23< vultraz_iOS> Tl'dr I didn't consider it might be useful for debugging and assumed it was an attempt to cap game performance
20170504 13:01:53-!- atarocch [~atarocch@64.251.121.244] has joined #wesnoth-dev
20170504 13:02:16-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has joined #wesnoth-dev
20170504 13:02:16-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has quit [Changing host]
20170504 13:02:16-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has joined #wesnoth-dev
20170504 13:16:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 13:17:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 13:25:01< irker576> wesnoth: Charles Dang wesnoth:master 7c89a984d700 / src/gui/core/event/ (distributor.cpp distributor.hpp): Fixup 039463e (don't use _t suffix) https://github.com/wesnoth/wesnoth/commit/7c89a984d700810020028fa41344e8b7b8c97ab7
20170504 13:40:59< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt f8d428ac68c5 / src/credentials.cpp: Use RAII for all that zero-filling https://github.com/wesnoth/wesnoth/commit/f8d428ac68c580e0521ea6f06cac06432402b1ec
20170504 13:41:01< irker576> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 0483bb762a4f / utils/travis/install_deps.sh: Attempt to fix Mac build (Scons checks CXXFLAGS not CPPFLAGS) https://github.com/wesnoth/wesnoth/commit/0483bb762a4ff953532f737e06f807706a276470
20170504 13:45:41-!- celmin|sleep is now known as celticminstrel
20170504 14:11:15< vultraz_iOS> wait, variable templates aren't even supported in VS2015? whaaaaa
20170504 14:11:50< celticminstrel> ???
20170504 14:11:57< celticminstrel> Oh, template variables.
20170504 14:12:15< celticminstrel> Well, it's not like we need them?
20170504 14:12:33< vultraz_iOS> just found it surprising
20170504 14:12:51< celticminstrel> Template variables don't really offer anything new.
20170504 14:13:24< celticminstrel> It just lets you do it in fewer lines of code.
20170504 14:13:46< celticminstrel> A template struct containing a static variable would have the same effect.
20170504 14:17:06-!- JyrkiVesterinen [~JyrkiVest@85-76-67-254-nat.elisa-mobile.fi] has quit [Quit: .]
20170504 14:38:09< celticminstrel> Did I break the MP test with my wesnoth.require thing?
20170504 14:38:52< celticminstrel> I don't think it should have affected it though...
20170504 14:39:35< celticminstrel> The test does require helper.
20170504 14:42:50< celticminstrel> vultraz_iOS: Cutscene theme?
20170504 14:43:27< vultraz_iOS> Later
20170504 14:44:04< celticminstrel> Why?
20170504 15:00:36-!- Kwandulin [~Kwandulin@p200300760F6D804EB83E53A4B2CCCA71.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170504 15:31:13-!- gfgtdf [~chatzilla@x4e36a3c9.dyn.telefonica.de] has joined #wesnoth-dev
20170504 15:49:18-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has quit [Ping timeout: 260 seconds]
20170504 15:51:20< celticminstrel> Oh yeah, vultraz_iOS, if you intend to release on 1.13.8, maybe you should hurry up and deal with the tiling issue?
20170504 15:51:48< celticminstrel> It's not good for a release to suddenly drop features, particularly now that you're tagging it Beta.
20170504 15:55:43-!- gfgtdf_ [~chatzilla@78.54.163.201] has joined #wesnoth-dev
20170504 15:58:24-!- gfgtdf [~chatzilla@x4e36a3c9.dyn.telefonica.de] has quit [Ping timeout: 258 seconds]
20170504 15:58:38-!- gfgtdf_ is now known as gfgtdf
20170504 16:16:28-!- chiec [~chiec@clients-xsf-98.upc.es] has joined #wesnoth-dev
20170504 16:18:42< celticminstrel> Should I document wesnoth.map_location now?
20170504 16:19:23< celticminstrel> DeFender1031: Also, before I do that, should I rename it? It feels little long to me, but I can't think of anything better other than maybe "wesnoth.location".
20170504 16:19:47< celticminstrel> (For reference, it contains functions for manipulating locations, such as "get_relative_dir".)
20170504 16:21:19-!- Kwandulin [~Kwandulin@p200300760F6D804EB83E53A4B2CCCA71.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170504 16:28:50< irker576> wesnoth: Celtic Minstrel wesnoth:master ab215a6d4473 / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/ab215a6d4473076f3a5c126966c54b48887afc88
20170504 16:30:11-!- sevu [~Unknown@p54857127.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170504 16:47:51-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170504 16:51:00< gfgtdf> celticminstrel: hmn 'location.get_relative_dir' might also be the name of a function that return the relative path of a directory.
20170504 16:51:12< celticminstrel> ...pffft.
20170504 16:51:20< celticminstrel> I suppose it could be.
20170504 16:51:52-!- JyrkiVesterinen [~JyrkiVest@87-100-253-156.bb.dnainternet.fi] has joined #wesnoth-dev
20170504 17:18:15-!- Kwandulin [~Kwandulin@p200300760F6D804EB83E53A4B2CCCA71.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170504 17:39:46-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170504 17:39:46< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Release Celtic Minstrel ab215a6: Update changelog Failed
20170504 17:39:46< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-326
20170504 17:39:51-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170504 17:46:10-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170504 17:46:10< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Debug Celtic Minstrel ab215a6: Update changelog Failed
20170504 17:46:10< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-326
20170504 17:46:14-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170504 18:17:05-!- sevu [~Unknown@p54857127.dip0.t-ipconnect.de] has quit [Quit: Verlassend]
20170504 18:24:37< irker576> wesnoth: Celtic Minstrel wesnoth:master db92e95e4589 / data/themes/editor.cfg: Editor: Remove "draw # of bitmaps" from toolbar https://github.com/wesnoth/wesnoth/commit/db92e95e4589f4523f469bb77776a35b83c667ba
20170504 18:41:32< zookeeper> it'd be nice if things like that could pop out of existence only when the available space/resolution warrants it
20170504 18:42:01< celticminstrel> I think themes do support resolutions in a similar manner to GUI2, so it might be possible.
20170504 18:42:13< zookeeper> i mean not automatically, but just so the element could have some minimum_width=1024 key or whatever.
20170504 18:42:31< zookeeper> and sure one could make several themes, but that immediately becomes awkward to maintain.
20170504 18:43:11< celticminstrel> I don't suppose that view is available in-game BTW? Some scenarios contain local terrain graphics IIRC, which you wouldn't be able to see in the editor.
20170504 18:48:10-!- chiec [~chiec@clients-xsf-98.upc.es] has quit [Remote host closed the connection]
20170504 18:55:53-!- gfgtdf_ [~chatzilla@x4e36a3c9.dyn.telefonica.de] has joined #wesnoth-dev
20170504 18:56:10-!- gfgtdf [~chatzilla@78.54.163.201] has quit [Ping timeout: 240 seconds]
20170504 18:56:18-!- gfgtdf_ is now known as gfgtdf
20170504 18:59:20< DeFender1031> celticminstrel, gimme a sec to catch up
20170504 18:59:42< celticminstrel> gfgtdf: I'm replacing placement=location_name with location_id=location_name.
20170504 18:59:46-!- Kwandulin [~Kwandulin@p200300760F6D804EB83E53A4B2CCCA71.dip0.t-ipconnect.de] has quit [Ping timeout: 264 seconds]
20170504 19:00:09< DeFender1031> celticminstrel, https://wiki.wesnoth.org/LuaWML/Reorganization#map
20170504 19:00:45< DeFender1031> If it's stuff that fits the same category as this stuff, then go for map.
20170504 19:03:12< celticminstrel> gfgtdf: distance_between(loc1, loc2) -> n, get_adjacent_tiles(loc) -> loc[6], get_direction(loc, dir) -> loc, tiles_adjacent(loc1, loc2) -> bool, get_relative_dir(loc1, loc2) -> dir, and a few more technical functions.
20170504 19:03:19< celticminstrel> Uh, I meant DeFender1031 there
20170504 19:04:23< celticminstrel> Oh hey, distance_between and get_adjacent_tiles already existed in helper. Huh.
20170504 19:07:22-!- atarocch [~atarocch@64.251.121.244] has quit [Ping timeout: 245 seconds]
20170504 19:10:32< zookeeper> celticminstrel, not available in-game, no. but scenario-specific rules are very rare and not much of a concern anyway when it comes to optimization.
20170504 19:13:30< celticminstrel> I don't even know what "get_in_bases_N_NE" means...
20170504 19:13:31< zookeeper> (and the coordinates and terrain codes toggles aren't either)
20170504 19:13:33< celticminstrel> ^basis
20170504 19:13:52< DeFender1031> celticminstrel, then yeah, .map
20170504 19:14:58< celticminstrel> "rotate_right_around_center" seems pretty descriptive but also overly long. Maybe it's fine though?
20170504 19:19:08< irker576> wesnoth: Jyrki Vesterinen wesnoth:large-gui-canvas e32bb5e24da2 / / (5 files in 3 dirs): Use std::unique_ptr to manage the layout and context in pango_text https://github.com/wesnoth/wesnoth/commit/e32bb5e24da204d6c592e2b91baf43029a33839b
20170504 19:23:09-!- atarocch [~atarocch@64.251.121.244] has joined #wesnoth-dev
20170504 19:29:01< celticminstrel> Wait, why does unique_ptr have a second template argument?
20170504 19:29:27< JyrkiVesterinen> It's the deleter.
20170504 19:30:29< celticminstrel> I thought that was only specified in the constructor.
20170504 19:30:35< celticminstrel> MInd you, I'm more used to shared_ptr...
20170504 19:31:15< gfgtdf> hmm sure that you don't have to take the function by &g_object_unref instead of just g_object_unref ?
20170504 19:31:34< JyrkiVesterinen> No, I'm not sure about that.
20170504 19:31:47< JyrkiVesterinen> It compiles in Visual Studio at least, though.
20170504 19:32:37< gfgtdf> JyrkiVesterinen: hm ok guess it it wokr for travis it's finje then.
20170504 19:35:37< celticminstrel> The & is totally optional when taking the address of a function.
20170504 19:35:40< gfgtdf> i wonder whether std::unique_ptr<PangoLayout, decltype([](void* obj){ g_object_unref(obj)})> layout_; would also work. (and then just dont apssa a secodn oarameter in the dtor)
20170504 19:35:48< celticminstrel> It is not however optional when taking the address of a non-static member function.
20170504 19:36:28< celticminstrel> gfgtdf: I was actually thinking unique_ptr<PangoThingamajig, void(*)(void*)> should work.
20170504 19:36:59< celticminstrel> gfgtdf: I guess you also don't have objections to the thing I said a little while ago?
20170504 19:37:19< gfgtdf> celticminstrel: i think mine (if it works) is faster since the compiler knows at compieltime what dtor function is called
20170504 19:37:24< gfgtdf> celticminstrel: what was that again?
20170504 19:37:33< celticminstrel> About placement=
20170504 19:37:55< gfgtdf> celticminstrel: hmm no objections i guess, although i wonder what the advantages are.
20170504 19:38:24< celticminstrel> gfgtdf: The advantates are that you can specify map_overwrite or map_passable in conjunction with a named location.
20170504 19:38:27< celticminstrel> ^advantages
20170504 19:39:29< celticminstrel> gfgtdf: Though, I was separating out overwrite/passable into separate keys, too, so in principle I suppose only one of the two changes is necessary to get that advantage.
20170504 19:43:29< gfgtdf> hmm seems lei lamdbas are not  allowed as template arguments.
20170504 19:48:22< celticminstrel> ...how do I cancel --no-gui again...
20170504 19:53:08< celticminstrel> Oh, --showgui
20170504 20:09:09< celticminstrel> Okay, so there are actually two completely different versions of get_relative_dir.
20170504 20:09:21< celticminstrel> One is called "default" and the other is called "radial symmetry".
20170504 20:09:31< celticminstrel> Naturally, the default one is "radial symmetry". 9_9
20170504 20:10:50< celticminstrel> map/location.cpp:255
20170504 20:15:04-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has joined #wesnoth-dev
20170504 20:15:04-!- deathisundead [~quassel@c-73-144-114-234.hsd1.mi.comcast.net] has quit [Changing host]
20170504 20:15:04-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has joined #wesnoth-dev
20170504 20:37:05< irker576> wesnoth: Jyrki Vesterinen wesnoth:large-gui-canvas afcca42abbbc / src/serialization/ (string_utils.cpp string_utils.hpp): Add utils::vertical_split() https://github.com/wesnoth/wesnoth/commit/afcca42abbbcca1df7bea29d2e36f7368a828f13
20170504 20:45:14-!- JyrkiVesterinen [~JyrkiVest@87-100-253-156.bb.dnainternet.fi] has quit [Quit: Going to bed]
20170504 20:52:15-!- sevu [~Unknown@141.39.226.226] has joined #wesnoth-dev
20170504 20:53:27< celticminstrel> Okay, so "default" breaks ties in favour of south...
20170504 20:53:49< celticminstrel> And "radially symmetric" breaks ties in favour of counter-clockwise...
20170504 20:57:02< celticminstrel> Presumably "default" was originally used for animations so that mreo units directly face the player, but "default" is actually used nowhere.
20170504 20:57:19< celticminstrel> Probably because someone changed the default from "default" to "radially symmetric".
20170504 21:00:01< celticminstrel> Apparently iceiceice changed it because "default" was somehow broken or something?
20170504 21:11:24< gfgtdf> celticminstrel: where is the default specified ?
20170504 21:11:55< celticminstrel> There's two overloads of get_relative_dir, the one-argument forwards to the two-argument with a default second argument.
20170504 21:12:25< celticminstrel> Changed in https://github.com/wesnoth/wesnoth/commit/cf19cc55162518a239f436c5e906889a7e5dd26f
20170504 21:12:59< celticminstrel> You can compare the results of the two with -ttest_relative_dir
20170504 21:13:12< gfgtdf> ok
20170504 21:13:17< celticminstrel> Moving the elf shows directions using "default", moving the orc shows directions using "radial symmetry".
20170504 21:13:39-!- shiki [~Unknown@141.39.226.226] has joined #wesnoth-dev
20170504 21:14:08< celticminstrel> IIUC it's calculating the relative dir from the dest location to every other location.
20170504 21:16:40-!- sevu [~Unknown@141.39.226.226] has quit [Ping timeout: 240 seconds]
20170504 21:16:56< celticminstrel> Should I pass weapons to the victory animation, or only to the death animation?
20170504 21:21:20< gfgtdf> celticminstrel: how is a weapoin involved in victory?
20170504 21:21:37< celticminstrel> gfgtdf: Well, I'm not sure, that's why I'm asking.
20170504 21:21:55< gfgtdf> celticminstrel: (assiming its a 'scenario won' animation not a 'unit killed' animation)
20170504 21:22:11< celticminstrel> It's unit victory.
20170504 21:22:34< celticminstrel> IIRC it's used by the ulfserker?
20170504 21:22:47< celticminstrel> Plays after they kill a unit.
20170504 21:25:08< celticminstrel> gfgtdf: Does that make a difference?
20170504 21:26:00< zookeeper> celticminstrel, uh, i don't understand the (original) question.
20170504 21:26:28< celticminstrel> Which question?
20170504 21:26:28< zookeeper> can't victory anims currently filter on the weapon(s) used?
20170504 21:26:39< celticminstrel> I don't know, that's why I'm asking.
20170504 21:26:53-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170504 21:27:16< celticminstrel> Thinking about it... the ulf victory is when the target has no melee attack, right? So I guess maybe they can?
20170504 21:28:09< zookeeper> none of this makes any sense to me.
20170504 21:28:10< celticminstrel> Ah, that wasn't a victory anim.
20170504 21:28:22< celticminstrel> It's instead a special attack animation.
20170504 21:28:37< zookeeper> why are you asking whether you should pass weapons to victory anims?
20170504 21:28:41-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170504 21:28:53< celticminstrel> For [kill]
20170504 21:28:59< irker576> wesnoth: Nils Kneuper wesnoth:master 1e02a0fcc3b5 / po/ (19 files in 19 dirs): updated Chinese (Simplified) translation https://github.com/wesnoth/wesnoth/commit/1e02a0fcc3b54f8ecbda8ce14dd9b3048bff932c
20170504 21:29:00< zookeeper> oh. okay.
20170504 21:30:05< celticminstrel> There are two victory animations in core, neither of which filter on weapons.
20170504 21:30:08-!- gfgtdf [~chatzilla@x4e36a3c9.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170504 21:30:28< zookeeper> well, depends on whether you also add a way to specify the weapons, i guess.
20170504 21:30:51< zookeeper> i don't know if you have
20170504 21:31:39< celticminstrel> On the other hand, [animate_unit] does pass weapons to victory animations. I should've checked there in the first place.
20170504 21:35:58-!- atarocch [~atarocch@64.251.121.244] has quit [Ping timeout: 268 seconds]
20170504 21:36:23< celticminstrel> It's actually a bit strange the way it uses a "dummy" attack for death animations... maybe someone should check sometime if that's actually needed...
20170504 21:44:22-!- Shiki_ [~Unknown@141.39.226.226] has joined #wesnoth-dev
20170504 21:47:28-!- shiki [~Unknown@141.39.226.226] has quit [Ping timeout: 240 seconds]
20170504 21:52:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 21:54:37-!- Shiki_ [~Unknown@141.39.226.226] has quit [Remote host closed the connection]
20170504 22:05:29-!- gfgtdf [~chatzilla@x4e3695a7.dyn.telefonica.de] has joined #wesnoth-dev
20170504 22:06:51-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20170504 22:12:42-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170504 22:13:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 22:22:46-!- atarocch [~atarocch@107.16.87.68] has joined #wesnoth-dev
20170504 22:30:40< vultraz_iOS> celticminstrel: what tiling issue
20170504 22:31:17< celticminstrel> Storyscreen
20170504 22:31:31< vultraz_iOS> meh
20170504 22:31:39< celticminstrel> Not meh, it needs to be don.
20170504 22:31:41< celticminstrel> ^done
20170504 22:31:47< celticminstrel> The lack is a blocker at least for 1.14.
20170504 22:32:09-!- mattiaskrgr is now known as matthiaskrgr
20170504 22:32:26< vultraz_iOS> meh
20170504 22:33:16< matthiaskrgr> should I do a woptipng run for 1.14 ? 
20170504 22:33:19< celticminstrel> Okay, there, not it's an actual issue so we don't forget.
20170504 22:33:22< celticminstrel> ^now
20170504 22:33:34< celticminstrel> Speaking of issues, did you see the one I added about...
20170504 22:33:40< celticminstrel> The editor?
20170504 22:33:43< celticminstrel> gfgtdf too
20170504 22:33:54< celticminstrel> 1023
20170504 22:33:56< vultraz_iOS> something about it
20170504 22:35:13< celticminstrel> I just closed all the milestones and added a due date to 1.13.8.
20170504 22:35:24< celticminstrel> That is, all the milestones except 1.13.8, obviously.
20170504 22:37:40< vultraz_iOS> I have no idea what the point of a border layout widget would be
20170504 22:38:03< celticminstrel> It has uses.
20170504 22:38:20< celticminstrel> In particular I think it might be useful for a GUI2 game UI.
20170504 22:38:27< vultraz_iOS> why?
20170504 22:38:37< celticminstrel> Because the game UI is pretty much like a border layout.
20170504 22:38:50< vultraz_iOS> what?
20170504 22:39:11< celticminstrel> Border layout is one of the basic layout manager types.
20170504 22:39:32< celticminstrel> Others are flow layout and grid layout.
20170504 22:39:39< celticminstrel> Wesnoth only supports grid layout currently.
20170504 22:39:48< vultraz_iOS> explain more 
20170504 22:40:08< celticminstrel> Flow layout is basically how you lay out a line of HTML text.
20170504 22:40:37< celticminstrel> You place elements one at a time, shifting to the right for each one, and start a new line when you run out of horizontal space.
20170504 22:40:59< celticminstrel> Border layout arranges components into five parts - the center, and each of the edges.
20170504 22:41:00-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Ping timeout: 260 seconds]
20170504 22:41:20< celticminstrel> Basically you have a pane that sticks to each edge, and a pane that takes up all the remaining space.
20170504 22:41:40< vultraz_iOS> (btw, what is this location list you speak of)
20170504 22:41:45< celticminstrel> Hm?
20170504 22:41:47< celticminstrel> Oh.
20170504 22:41:52< celticminstrel> Click the flag tool.
20170504 22:41:58< vultraz_iOS> celticminstrel: and you just described the matrix widget 
20170504 22:42:02< celticminstrel> Add a location, call it whatever you want, and place it on the map.
20170504 22:42:14< celticminstrel> Save the map, close it, reload.
20170504 22:42:39< celticminstrel> Click the flag tool. Notice that the location is marked on the map but not listed in the palette, even when you scroll down to the bottom.
20170504 22:42:57< celticminstrel> Yes, I think the matrix widget was intended to be a border layout, but you said it doesn't actually work.
20170504 22:42:57-!- Alkenrinnstet [~alkenrinn@42.61.217.253] has quit [Ping timeout: 252 seconds]
20170504 22:43:09< vultraz_iOS> i don't know
20170504 22:43:16< vultraz_iOS> i haven't really used it
20170504 22:43:24< vultraz_iOS> i tried it once for a weird thing and couldn't get it to work
20170504 22:43:25< celticminstrel> ISTR you tried to though.
20170504 22:43:30< celticminstrel> I see.
20170504 22:43:40< vultraz_iOS> IIRC it's used in the debug clock?
20170504 22:44:10< celticminstrel> Was it? I don't recall.
20170504 22:44:21< celticminstrel> So maybe it does actually work?
20170504 22:44:36< celticminstrel> The matrix does have some weird stuff in it though, like that delayed instantiation or whatever.
20170504 22:45:16< vultraz_iOS> I don't know I don't know I don't know!
20170504 22:45:36< celticminstrel> You don't need to say it thrice.
20170504 22:48:58-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20170504 22:50:35-!- gfgtdf [~chatzilla@x4e3695a7.dyn.telefonica.de] has quit [Ping timeout: 268 seconds]
20170504 22:54:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170504 22:54:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170504 22:56:05-!- gfgtdf [~chatzilla@x4e363c4a.dyn.telefonica.de] has joined #wesnoth-dev
20170504 22:56:27-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds]
20170504 22:57:08< gfgtdf> vultraz_iOS: we also have some blockers in the tracker, most of them are multiplayer issues
20170504 22:57:16< vultraz_iOS> yes
20170504 22:57:21< vultraz_iOS> i can't do anything about that
20170504 22:58:00< celticminstrel> I was going to work more on the issues porting today but kinda forgot, huh...
20170504 22:58:04< gfgtdf> vultraz_iOS: still we cannot really a relase a rc version with that.
20170504 22:58:06< celticminstrel> Well, there's still time though...
20170504 22:58:14< celticminstrel> Need to decide what to do with the attachments.
20170504 22:58:58< gfgtdf> vultraz_iOS: we also have this issue http://gna.org/bugs/index.php?25467 from which we don't wknoe whether it's still present on master, so we'll just postpone ot until after 1.13.8 and test whetehr it's still in 1.13.8
20170504 23:01:50< gfgtdf> vultraz_iOS: on https://gna.org/bugs/?25685 i ahve an idea why coudl be the casue but the onyl fis is something i originally wanted to avoid since it included adding a worker thread object in the wesnothd_connection class.
20170504 23:02:15< gfgtdf> vultraz_iOS: i have no idea on https://gna.org/bugs?25686 though ?
20170504 23:02:44< gfgtdf> vultraz_iOS: wedge009 said a he couldn'T reprpoduce it, can you now come to the 1.13.7 mp server to try to reproduce it?
20170504 23:03:19< gfgtdf> vultraz_iOS: with a master client i meant
20170504 23:03:26< vultraz_iOS> alright
20170504 23:06:55< gfgtdf> vultraz_iOS: tell me when you there
20170504 23:07:08< vultraz_iOS> yeah im here
20170504 23:07:43< gfgtdf> vultraz_iOS: hmm i don't se you, sure aou ar eon the 1.13.7 server ?
20170504 23:08:04< gfgtdf> that the port 14997 server
20170504 23:08:07< vultraz_iOS> oh
20170504 23:08:07< vultraz_iOS> no
20170504 23:08:58< gfgtdf> pls join
20170504 23:10:03< gfgtdf> accidently stil has addon ables have to restert game
20170504 23:10:41< gfgtdf> vultraz_iOS: ^ pls join the other game
20170504 23:13:15< vultraz_iOS> gfgtdf: i wrote text are you not seeing it
20170504 23:13:45< gfgtdf> vultraz_iOS: no i'm not, seems to e bug on th ereceiving then that was probably fixed since 1.13.7 release will look more into it
20170504 23:13:57< gfgtdf> reveiving end*
20170504 23:14:05< vultraz_iOS> ok
20170504 23:20:16-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Read error: Connection reset by peer]
20170504 23:21:44-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20170504 23:48:29-!- sharpe [~quassel@cpc105720-shef14-2-0-cust238.17-1.cable.virginm.net] has joined #wesnoth-dev
20170504 23:48:42< sharpe> hello guys :)
20170504 23:49:24< vultraz_iOS> hello
20170504 23:50:04< sharpe> i'm looking to get involved with contributing to wesnoth
20170504 23:50:12< sharpe> :)
20170504 23:53:01< sharpe> just finished uni so need to fill some time. i'm a pretty experienced c++ programmer. I read the getting started thread on the forum and i was wondering how i'd pick something to work on. Is there a list of stuff anywhere that noone is working on yet? :)
20170504 23:55:30< gfgtdf> sharpe: sure, we have the bugtracker and the easycoding wiki page
20170504 23:56:07< gfgtdf> sharpe: you have a particular part that you want to work on ?
20170504 23:59:42< celticminstrel> Oh hi.
20170504 23:59:54< celticminstrel> Yay more contributors.
20170504 23:59:59< Aginor> :)
--- Log closed Fri May 05 00:00:25 2017