--- Log opened Fri Jun 05 00:00:06 2015
20150605 00:08:40-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 00:11:03-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20150605 00:32:08-!- cib0 [~cib@p5DC755DA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 00:40:45-!- Appleman1234 [~Appleman1@s108.BMT-e1.vectant.ne.jp] has quit [Remote host closed the connection]
20150605 00:53:34-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 00:56:03-!- cib0 [~cib@p5DC755DA.dip0.t-ipconnect.de] has quit [Ping timeout: 256 seconds]
20150605 00:58:34-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 265 seconds]
20150605 01:12:26-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20150605 01:23:01-!- irker321 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150605 01:23:01< irker321> wesnoth: Ignacio R. Morelle wesnoth:master 7ec2dc015e5b / changelog players_changelog src/gui/dialogs/chat_log.cpp: gui2/tchat_log: Start dialog displaying the last chat log page http://git.io/vIke8
20150605 01:23:02< irker321> wesnoth: Ignacio R. Morelle wesnoth:master 98bcaa784e23 / changelog data/core/music/silence.ogg players_changelog: Use a larger silence.ogg to work around bug #23633 http://git.io/vIke4
20150605 01:23:03< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 292b09e55a22 / changelog players_changelog src/gui/dialogs/chat_log.cpp: gui2/tchat_log: Start dialog displaying the last chat log page http://git.io/vIkeB
20150605 01:23:05< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 a26de26fdd6f / changelog data/core/music/silence.ogg players_changelog: Use a larger silence.ogg to work around bug #23633 http://git.io/vIkeR
20150605 01:25:08< shadowm> gui2::tscrollbar_container::scroll_vertical_scrollbar() appears to do nothing at all, at least when applied to a scroll_label. :\
20150605 01:30:32-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20150605 01:53:57-!- Greg-Boggs_ [~quassel@173.240.247.62] has quit [Ping timeout: 244 seconds]
20150605 01:57:05< shadowm> gfgtdf: f16dec91b93f08844d999c9056d165951c980148 isn't a very good fix, the correct fix is to make that int a reference.
20150605 01:57:44< shadowm> And yes, I noticed the parameter is never used in the caller. Chalk it up to yet another unfinished editor feature.
20150605 01:58:08-!- ancestral [~ancestral@65.103.33.158] has joined #wesnoth-dev
20150605 01:58:53< shadowm> I don't feel like fixing it right now so I'll just wait for someone to file a bug report.
20150605 02:05:00< shadowm> ancestral: Still haven't had time to test that patch? :\
20150605 02:05:40< shadowm> Did you see my email? Will you have time to package those releases or should I contact mattsc instead?
20150605 02:08:31< irker321> wesnoth: Ignacio R. Morelle wesnoth:master fe27196bed95 / src/gui/dialogs/editor/ (editor_edit_side.cpp editor_edit_side.hpp): gui2/teditor_edit_side: Make controller_num ctor parameter a reference http://git.io/vIkqN
20150605 02:08:34< irker321> wesnoth: Ignacio R. Morelle wesnoth:master e817858cbb21 / changelog players_changelog: Add changelog entries for last commit http://git.io/vIkqA
20150605 02:08:37< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 dd8b36fb16c4 / src/gui/dialogs/editor/ (editor_edit_side.cpp editor_edit_side.hpp): gui2/teditor_edit_side: Make controller_num ctor parameter a reference http://git.io/vIkqh
20150605 02:08:40< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 518f6fe9b636 / changelog players_changelog: Add changelog entries for last commit http://git.io/vIkqj
20150605 02:10:52-!- ancestral [~ancestral@65.103.33.158] has quit [Ping timeout: 272 seconds]
20150605 02:12:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150605 02:12:59-!- travis-ci [~travis-ci@ec2-54-146-146-56.compute-1.amazonaws.com] has joined #wesnoth-dev
20150605 02:13:00< travis-ci> wesnoth/wesnoth#6574 (master - 98bcaa7 : Ignacio R. Morelle): The build was fixed.
20150605 02:13:00< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/65503419
20150605 02:13:00-!- travis-ci [~travis-ci@ec2-54-146-146-56.compute-1.amazonaws.com] has left #wesnoth-dev []
20150605 02:14:19< shadowm> Blargh stupid GitHub PR list won't let me filter by author.
20150605 02:14:36< shadowm> Oh wait, it's because I got the name wrong and it won't warn me.
20150605 02:19:06< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 5149601c9303 / src/game_controller.cpp: Improve formatting of the MP save/replay warning for Load Game http://git.io/vIkOD
20150605 02:19:53-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has quit [Remote host closed the connection]
20150605 02:21:32-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20150605 02:26:10< irker321> wesnoth: Elvish_Hunter wesnoth:1.12 7339dc9a121e / changelog data/multiplayer/factions/rebels-aoh.cfg players_changelog: Removed Silver Mage from the Rebels' leaders in Age of Heroes http://git.io/vIksG
20150605 02:41:45-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 02:46:32-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 252 seconds]
20150605 03:36:41< irker321> wesnoth: Ignacio R. Morelle wesnoth:master b4b7a3656fad / / (4 files in 2 dirs): mp: Implement whisper_friends_only option in GUI1 lobby with a periodic warning http://git.io/vIkz9
20150605 03:36:46< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 7be7e3ce8cf8 / changelog players_changelog src/multiplayer_ui.cpp src/multiplayer_ui.hpp: mp: Implement whisper_friends_only option in GUI1 lobby with a periodic warning http://git.io/vIkzQ
20150605 03:44:59-!- SpoOkyMagician [~chatzilla@cpe-74-136-81-20.kya.res.rr.com] has quit [Quit: meh]
20150605 04:29:59-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 04:34:47-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 256 seconds]
20150605 04:41:40-!- {V} [~V@105-70-ftth.on.nl] has quit [Read error: Connection reset by peer]
20150605 04:42:07-!- {V} [~V@105-70-ftth.on.nl] has joined #wesnoth-dev
20150605 05:04:15-!- oldlaptop [~quassel@50-107-82-76.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 248 seconds]
20150605 05:13:10-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 05:13:56-!- Appleman1234 [~Appleman1@s108.BMT-e1.vectant.ne.jp] has joined #wesnoth-dev
20150605 05:15:08-!- oldlaptop [~quassel@50-107-82-76.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150605 05:39:27-!- shadowm changed the topic of #wesnoth-dev to: 1.12 branch string freeze ETA: June 5th 20:00 UTC-03 | Wesnoth 1.12.3 ETA: June 12th 20:00 UTC-03 | Wesnoth 1.13.1 ETA: TBD | Wesnoth Developers Channel | Logs: http://irclogs.wesnoth.org
20150605 05:45:02-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has joined #wesnoth-dev
20150605 05:45:20< ancestral> shadowm: Tried your patch
20150605 05:45:23< ancestral> I’m getting build errors, sadly
20150605 05:45:34< shadowm> Paste.
20150605 05:47:05< ancestral> https://paste.ee/p/chGyP
20150605 05:47:12< ancestral> It’s possible it might not be your patch that caused it
20150605 05:47:17< ancestral> I am building against master
20150605 05:47:29< shadowm> It's pretty obvious it's not my patch.
20150605 05:47:39< ancestral> Okay
20150605 05:47:51< ancestral> I could go back, or I can try to debug this thing
20150605 05:48:08< ancestral> Looks like more library garbage
20150605 05:48:20< shadowm> src/make_enum.cpp:7:    void debug_conversion_error(const std::string& temp, const bad_enum_cast & e)
20150605 05:48:38< shadowm> Make sure your project compiles that file.
20150605 05:48:59< ancestral> Or I could be wrong
20150605 05:50:48< ancestral> Aha new file
20150605 05:50:51< shadowm> In general, if you stumble upon an error like this you just need to grep the source for the definition of the offending symbol, usually in a .cpp file.
20150605 05:51:13< shadowm> Then make sure the file in question is part of the project, compiled, and linked into the target executable.
20150605 05:59:00< ancestral> Added those files
20150605 05:59:03< ancestral> Build succeeded
20150605 06:02:30-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has quit [Read error: Connection reset by peer]
20150605 06:02:52-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has joined #wesnoth-dev
20150605 06:03:30< shadowm> 20150601 04:31:12< shadowm> Then run Wesnoth with --log-info=font
20150605 06:03:37< shadowm> ancestral: ^
20150605 06:03:46< ancestral> https://paste.ee/p/y9nLN
20150605 06:03:54< shadowm> 20150601 04:47:50< shadowm> ancestral: Okay, and the visual result?
20150605 06:04:02< ancestral> I have a library issue that is causing a segmentation fault
20150605 06:04:07< shadowm> God dammit.
20150605 06:04:07< ancestral> I need to investigate that
20150605 06:04:09< ancestral> I know
20150605 06:04:20< ancestral> I am as frustrated as you are
20150605 06:04:28< shadowm> Then we literally achieved nothing we didn't already achieve on June 1st. :p
20150605 06:04:42< shadowm> Also,
20150605 06:04:44< shadowm> 23:05:42 <shadowm> Did you see my email? Will you have time to package those releases or should I contact mattsc instead?
20150605 06:04:49< ancestral> This last week has been ridiculously busy
20150605 06:05:07< ancestral> I have 3 days off this weekend
20150605 06:05:08< shadowm> (and/or postpone everything)
20150605 06:05:12< ancestral> I have an additional day off on the 13th
20150605 06:05:41< shadowm> Okay. You *will* package 1.12.3 and potentially 1.13.1 that day.
20150605 06:06:03< ancestral> So… I can confidently release 1.12.3
20150605 06:06:05< shadowm> Otherwise I'll either contact mattsc or postpone everything.
20150605 06:06:18< ancestral> 1.13.1 is where there are unsolved library issues
20150605 06:06:47< ancestral> If mattsc is willing to help out with 1.13.1, I wouldn’t mind
20150605 06:07:20< ancestral> So yes, for 1.13.1, I would appreciate it
20150605 06:08:45< shadowm> I'll ask him what he thinks about this then.
20150605 06:10:06< shadowm> loonycyborg: I've yet to hear from you regarding the libogg/libvorbis* bugs or whether you'll be able to package the Windows installer that day.
20150605 06:11:27< shadowm> loonycyborg: I'd appreciate an answer before 20:00 UTC-03 so 1.12 doesn't spend extra time frozen for nothing in the worst case.
20150605 06:13:11< ancestral> shadowm: k
20150605 06:13:15< ancestral> I got it to launch
20150605 06:13:25< ancestral> I had to copy over some pango files, and that’s it
20150605 06:13:39< shadowm> Okay, great. What font is in use?
20150605 06:14:05< ancestral> For the hints/tips, Helvetica :(
20150605 06:14:28< shadowm> Everything on the title screen uses a single font. It's all Helvetica still?
20150605 06:14:55< shadowm> And did you get the same output as before with --log-info=font?
20150605 06:15:12< shadowm> (I only care about the "info font:" lines.)
20150605 06:15:35< shadowm> (Specifically, those before the "Data directory" line.)
20150605 06:15:44< ancestral> http://i.imgur.com/Q6goO4O.png
20150605 06:15:46< ancestral> One second looking
20150605 06:16:18< ancestral> Yeah, it’s the same
20150605 06:16:20< shadowm> Yeah, I don't think that's DV Sans.
20150605 06:16:28< shadowm> Welp.
20150605 06:16:28< ancestral> Andagii to DroidSansJapanese
20150605 06:16:36< ancestral> The screenshot shows the story on the G
20150605 06:16:41< ancestral> It’s the easiest way to tell imo
20150605 06:16:49< ancestral> But the buttons are DejaVu I believe
20150605 06:16:59< shadowm> That's physically impossible.
20150605 06:17:04< ancestral> Is it?
20150605 06:17:06< ancestral> I could be wrong
20150605 06:17:18< shadowm> All of GUI2 uses the same font except for three dialogs.
20150605 06:17:35< ancestral> Preferences is not GUI2, yes?
20150605 06:17:42< ancestral> s/yes/correct
20150605 06:17:46< shadowm> No, it's not.
20150605 06:17:59< shadowm> But all of the title screen is one big fullscreen GUI2 dialog.
20150605 06:18:12-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 06:18:25 * ancestral looks more closely
20150605 06:18:37< ancestral> Yeah, okay, it’s different screens where the buttons are DejaVu
20150605 06:18:58< ancestral> The text is slightly smaller and narrower in the buttons in comparison
20150605 06:22:04< shadowm> Welp, I'm out of ideas now because I can't find anything on the web suggesting some OS X-specific fontconfig bug being in play.
20150605 06:22:52-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 256 seconds]
20150605 06:24:04< ancestral> I wonder if anyone else building using a SDL/ttf/fontconfig/pango/cairo stack is running into these issues too
20150605 06:24:26< ancestral> and has an answer
20150605 06:24:40 * ancestral is thinking of projects, not individuals
20150605 06:24:43< shadowm> I honestly doubt there's anyone else as stupid as us to mix SDL_ttf with the Pango/Cairo pipeline in the same application.
20150605 06:25:14< shadowm> Well, not mix, but use in parallel.
20150605 06:25:54< ancestral> shadowm: Honestly, I’ve noticed Helvetica shown in GUI2 ever since GUI2 was implemented
20150605 06:26:09< ancestral> It’s become more apparent as windows have migrated that way
20150605 06:27:11< shadowm> Hm.
20150605 06:27:19< ancestral> I can’t remember; maybe it did at one time work with having DejaVu Sans installed in the system
20150605 06:27:23< ancestral> Which isn’t happening now
20150605 06:27:56< shadowm> I wouldn't be surprised if that helped.
20150605 06:28:14< shadowm> The problem is that fontconfig refuses or can't use Wesnoth's local fonts dir on OS X for some reason.
20150605 06:28:21< ancestral> I do have “DejaVu Sans” installed
20150605 06:28:21< shadowm> I mean, probably.
20150605 06:28:25-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 06:28:33< shadowm> Eh, then I don't know.
20150605 06:29:41< ancestral> I guess we’re fortunate it’s using another grotesque sans-serif font and not, say, something like Times or whatever
20150605 06:29:56-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 06:30:09< ancestral> So it looks similar (even if it is off and odd-looking in some cases)
20150605 06:30:49< shadowm> Do you have an older screenshot lying around?
20150605 06:31:48< shadowm> Say, from before commit 5a557bfecabae5d76c5dc753caf265cab52f0891 (June 1 02:09:03 UTC-03).
20150605 06:32:00< shadowm> I'm just curious about a tiny inconsequential detail unrelated to this.
20150605 06:32:47< ancestral> I’m sure I do
20150605 06:34:56< ancestral> Of the main screen?
20150605 06:35:25< ancestral> Hmm
20150605 06:35:41< shadowm> Anything GUI2, really.
20150605 06:35:59< shadowm> If not then don't worry.
20150605 06:36:59< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 fbe7a58c6f06 / src/md5.cpp: md5: Return a NULL pointer if MD5::raw_digest() is called early http://git.io/vIILq
20150605 06:37:02< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 1dd7a4614489 / README README.md: Convert our readme file to Markdown http://git.io/vIILm
20150605 06:37:05< irker321> wesnoth: Ignacio R. Morelle wesnoth:1.12 5d993718a67b / .travis.yml: travis: Use -pipe to pipe compiler output directly into the assembler http://git.io/vIILY
20150605 06:37:33< ancestral> http://imgur.com/akQqun6
20150605 06:37:42< ancestral> 2 months ago
20150605 06:38:14< shadowm> It looks *that* awful with the smaller font size?
20150605 06:38:21< ancestral> Yeah :(
20150605 06:38:33< shadowm> Well, that's not GUI2, but it uses ttext.
20150605 06:38:47< ancestral> If I had more time, I’d create a theme with larger text
20150605 06:38:49< shadowm> So, good enough since GUI2 uses ttext.
20150605 06:39:07< shadowm> fabi chose to use a smaller font for vresolutions < 768 in order to save space.
20150605 06:39:25< ancestral> shadowm: It’s possible I took that not in standard resolution but native resolution; i.e. not pixel double. (Which means it’d be larger, but look crummier because the anti-aliasing gets blown up too)
20150605 06:39:35< shadowm> So far this has only served to confuse people and get false positives reported ini the forums.
20150605 06:40:04< ancestral> I know it’s possible to serve up larger fonts dependent on the screen size
20150605 06:40:09< shadowm> So, I'm tempted to revert that bit.
20150605 06:40:47< ancestral> A month ago, but I don’t know if this is helpful: http://i.imgur.com/JBG5EBy.png
20150605 06:41:37< shadowm> GUI0, not using ttext there.
20150605 06:41:38< ancestral> One more: http://imgur.com/N2ISz4h a month ago
20150605 06:42:05< shadowm> That's GUI0 using ttext.
20150605 06:42:09< ancestral> Hmm okay
20150605 06:42:19< shadowm> Okay, thanks, that one is all I needed to know.
20150605 06:42:22< ancestral> Cool
20150605 06:43:00< shadowm> So I'm curious why Pango/Cairo aren't enabling subpixel hinting in Wesnoth on OS X pre-5a557bfecabae5d76c5dc753caf265cab52f0891 even though the window decoration text is clearly subp-hinted.
20150605 06:43:22< ancestral> I’ll get in touch with matt over the next few days and heopfully we can try swapping  builds of master (see if mine works for him and vice versa)
20150605 06:43:29< shadowm> Maybe it's connected to their inability to use a non-default font.
20150605 06:43:48< ancestral> shadowm: So about that
20150605 06:43:56< ancestral> The window decoration hinting
20150605 06:44:10< ancestral> Building through Homebrew, it works, looks “more retina”
20150605 06:44:16< ancestral> Building through Xcode? No
20150605 06:44:30< ancestral> My guess it’s either a flag or difference with libraries
20150605 06:44:46< shadowm> Does "more retina" mean "uses color subpixel hinting?
20150605 06:44:56< ancestral> Well
20150605 06:45:19< ancestral> I thought it was basically if the window had “retina support”
20150605 06:45:41< ancestral> In other words, the window title appears at pixel double with a retina resolution
20150605 06:45:57< ancestral> Versus, drawn at a higher DPI instead
20150605 06:46:07< shadowm> No, I'm only interested in the font antialiasing technique used.
20150605 06:46:13< ancestral> Oh, hmm
20150605 06:46:19< ancestral> I haven’t paid enough attention to that
20150605 06:46:28< shadowm> If you zoom into the window decoration, you'll see a lot of random colors in the font glyph antialiasing.
20150605 06:46:39< shadowm> Those are indicative of subpixel hinting taking place.
20150605 06:46:43< ancestral> Although I have not been overly impressed by the anti-alising techniques used by Wesnoth
20150605 06:46:53< ancestral> (Obviously, upstream issues)
20150605 06:46:54< shadowm> People who are not me don't have a problem with subpixel hinting.
20150605 06:47:01< shadowm> People who are me see the text bleeding colors.
20150605 06:47:12< shadowm> But.
20150605 06:47:25< shadowm> Wesnoth CAN use subpixel hinting on Linux, but it doesn't handle it well.
20150605 06:47:33< ancestral> And I think a good amount of that is magnified when pixel double blows it up
20150605 06:47:37< shadowm> As a result, text bleeds even more colors than it's supposed to do.
20150605 06:47:52< shadowm> So I decided to disable this forcefully on all non-Windows platforms with commit 5a557bfecabae5d76c5dc753caf265cab52f0891.
20150605 06:48:01< ancestral> I see
20150605 06:49:46< shadowm> AFAICT from these screenshots, this makes no difference on OS X because Wesnoth wasn't allowed to use subp hinting there in the first place for some weird reason.
20150605 06:50:17-!- boucman_work [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20150605 06:50:17-!- boucman_work [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Changing host]
20150605 06:50:17-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150605 06:50:33< shadowm> So I need not fear the OS X crowd coming to my front door with pitchforks and torches.
20150605 06:51:41< shadowm> Anyway, that's all for now. I'm off.
20150605 06:52:20< shadowm> I'm curious if this Helvetica nonsense actually benefits the OS X crowd.
20150605 06:53:00< shadowm> In DejaVu Sans, the uppercase I and lowercase l are identical except in height (meaning most mortals won't notice the difference without zooming in).
20150605 06:53:13< shadowm> That's uppercase i and lowercase L.
20150605 06:54:38< shadowm> Hm. Meh.
20150605 06:54:46< shadowm> Helvetica has the same issue AFAICT.
20150605 06:55:36< shadowm> Well, in that case we can't even replace DejaVu Sans with a less abusable font and expect it to work reliably on OS X. <_<
20150605 06:58:07< ancestral> Unless we (or uh, someone?) changes how text renders
20150605 06:58:22< ancestral> to some other way
20150605 06:58:38< shadowm> I doubt it's anyone's fault but Pango/Cairo/fontconfig's.
20150605 06:59:32< shadowm> And that's really our best option for cross-platform text rendering right now.
20150605 07:00:34< shadowm> ttext's font selection mechanism isn't really all that complicated.
20150605 07:01:13< shadowm> We have a list of fonts, we stuff that list of fonts into Pango, Pango gives us a handle of sorts to the font we'll use.
20150605 07:02:55< shadowm> The rest is basically asking Pango to render on a Cairo surface and decoding the result.
20150605 07:03:08< shadowm> The result being the rasterized text.
20150605 07:04:32< shadowm> The font files to load are determined during startup by registering the fonts dir with fontconfig (X11 and OS X) or informing the Win32 API of the existence of the individual files (Windows).
20150605 07:05:35< shadowm> So by the point we call Pango or Cairo, fontconfig (their font provider) is supposed to know which font families map to which files.
20150605 07:18:32< ancestral> shadowm: So, on a basic level, what is GUI 1 doing that GUI 2 doesn’t that lets Wesnoth render the main font correctly?
20150605 07:18:58< ancestral> Or am I off, it’s not a GUI 1 vs. GUI 2 thing
20150605 07:19:34< shadowm> GUI1 uses SDL_ttf, which uses FreeType directly to both read font files and render text.
20150605 07:21:16< shadowm> GUI2 uses our ttext wrapper class, which uses Pango to render text using Cairo, which may or may not use FreeType or some other platform-specific backend. Cairo uses fontconfig to load system-wide or application-specific font files if necessary.
20150605 07:22:15< shadowm> (Cairo on OS X will use whatever OS X's native font API is, I believe. On Windows it'll use GDI instead of FreeType. On Linux, FreeType is the only possible font API.)
20150605 07:22:25-!- [Relic] [~Relic]@2602:306:33a3:6d30:618c:545d:6cae:b943] has quit [Quit: Leaving]
20150605 07:24:23< shadowm> GUI1 doesn't use SDL_ttf directly, but rather does so through a few Wesnoth-specific functions that provide services such as render caching and font fallback.
20150605 07:24:59< shadowm> But for most intents and purposes and for the sake of simplicity we can pretend those functions don't exist.
20150605 07:26:57< shadowm> But things are more complicated than GUI1 vs. GUI2, because even if GUI2 didn't exist, GUI1 is *not* a consolidated framework.
20150605 07:27:32< shadowm> (GUI2 was created precisely to solve that. The result is https://xkcd.com/927/ .)
20150605 07:28:11< shadowm> The loading screen with the Wesnoth logo on a black background is not a GUI1 "thing", even though is leverages the same font rendering code paths as GUI1.
20150605 07:29:06< shadowm> On the flip side, story screens use GUI1 widgets for the buttons, and for the buttons only. The story screen title and content text blocks use ttext, just like GUI2.
20150605 07:30:37< shadowm> And then there's the in-game themable UI (which I call "theme UI" for short) with the sidebar and menu bar and minimap and main map viewport.
20150605 07:31:13< shadowm> The theme UI uses GUI1 widgets, but like the story screen, it uses ttext for some static text parts.
20150605 07:31:34< shadowm> IIRC the main map viewport uses ttext for rendering map labels as well.
20150605 07:33:20< shadowm> Yes, this is confusing as hell, isn't it?
20150605 07:33:53< shadowm> The default MP lobby uses GUI1 widgets and GUI1's font rendering code exclusively.
20150605 07:37:35< shadowm> So, because of a major bug which upstream doesn't seem interested in fixing despite my overly detailed bug report, I want to get rid of the SDL_ttf dependency.
20150605 07:38:04< shadowm> When I get around to doing that (it's not trivial) we'll only have the Pango/Cairo/fontconfig pipeline.
20150605 07:42:42-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150605 07:46:28-!- horrowind [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150605 07:52:52-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150605 07:56:35-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 08:00:59-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20150605 08:06:26-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 08:11:06-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20150605 08:11:07-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 244 seconds]
20150605 08:18:57-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20150605 08:20:56-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 252 seconds]
20150605 08:21:16-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 08:30:59-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 08:33:52-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 08:46:07-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150605 08:59:49-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150605 09:08:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150605 09:31:22-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20150605 09:36:13-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 265 seconds]
20150605 09:37:16-!- irker321 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150605 09:54:42-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 09:59:57-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 276 seconds]
20150605 10:38:18-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20150605 10:40:19-!- gfgtdf [~chatzilla@x50abb053.dyn.telefonica.de] has joined #wesnoth-dev
20150605 10:41:24< gfgtdf> shadowm: : idk what controller num does so i cannot fix it. and with that commit at least the reso the the edit side dialog seems to work again.
20150605 10:42:12-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 276 seconds]
20150605 10:44:02-!- gfgtdf [~chatzilla@x50abb053.dyn.telefonica.de] has quit [Client Quit]
20150605 10:51:31-!- Necrosporus_ is now known as Necrosporus
20150605 11:14:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20150605 11:19:12-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150605 11:19:52-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has joined #wesnoth-dev
20150605 11:20:40-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150605 11:42:53-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 11:47:36-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 244 seconds]
20150605 11:48:43-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 11:51:50-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 264 seconds]
20150605 12:02:52-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20150605 12:05:01-!- vultraz_iOS [uid24821@gateway/web/irccloud.com/x-fmtqulwrwqhhjmum] has joined #wesnoth-dev
20150605 12:05:11-!- vultraz_iOS [uid24821@gateway/web/irccloud.com/x-fmtqulwrwqhhjmum] has quit [Changing host]
20150605 12:05:11-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20150605 12:50:22-!- gfgtdf [~chatzilla@x50abb053.dyn.telefonica.de] has joined #wesnoth-dev
20150605 12:51:21< gfgtdf> vultraz_iOS: about: http://gna.org/bugs/?23630 maybe this can be fixed by implemting [role] in lua
20150605 12:54:28< vultraz_iOS>  That would likely work
20150605 12:54:40< vultraz_iOS> Since we can use split()
20150605 12:56:07< gfgtdf> vultraz_iOS: or we coudl just apply teh filter normally than then order teh resuted unit attay by type, idk whihc would be faster
20150605 12:58:34< vultraz_iOS> I think we should stop treating [role] type= as special and just use it as a standard suf key
20150605 12:59:03< gfgtdf> vultraz_iOS: hm that could brask some wml code.
20150605 12:59:10< vultraz_iOS> Why?
20150605 12:59:32< gfgtdf> vultraz_iOS: becasue some wml code might want texaclt ythis beahaviour
20150605 12:59:52< vultraz_iOS> There's not really any different IMO
20150605 12:59:57< vultraz_iOS> Just the implementation
20150605 13:00:28< gfgtdf> vultraz_iOS: it is deifferent with the new [role] different unit will get the role
20150605 13:00:46< vultraz_iOS> How so
20150605 13:01:11< gfgtdf> vultraz_iOS: wuith ccurrently type=type1,type and there are units of both types than its guarnteed that the type1 unit is selected
20150605 13:04:34< gfgtdf> vultraz_iOS: 
20150605 13:05:11< vultraz_iOS> Hmmm
20150605 13:05:24< vultraz_iOS> True
20150605 13:31:09-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 13:35:38-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 246 seconds]
20150605 13:52:52-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds]
20150605 13:53:19-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 14:01:01-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 264 seconds]
20150605 14:27:48-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 14:43:53-!- OnkV [~quassel@85.124.51.57] has joined #wesnoth-dev
20150605 15:08:58-!- Elvish_Hunter [~irssi@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20150605 15:09:26< Elvish_Hunter> Hi all
20150605 15:10:07< Elvish_Hunter> gfgtdf: did you look for me?
20150605 15:12:44-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20150605 15:19:25-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 15:21:23-!- Greg-Boggs [~quassel@173.240.247.62] has joined #wesnoth-dev
20150605 15:24:04-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 252 seconds]
20150605 15:30:06< gfgtdf> Elvish_Hunter: hm yes but i curretnly cannot rmember why
20150605 15:32:57-!- Crendgrim [~crend@wesnoth/forum-moderator/crendgrim] has quit [Remote host closed the connection]
20150605 15:33:15-!- Crendgrim [~crend@wesnoth/forum-moderator/crendgrim] has joined #wesnoth-dev
20150605 15:35:05-!- irker090 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150605 15:35:05< irker090> wesnoth: gfgtdf wesnoth:master f69481d26f6d / src/ (make_enum.hpp tests/test_make_enum.cpp): add parse method and static name method to MAKE_ENUM http://git.io/vIqcZ
20150605 15:35:05< irker090> wesnoth: gfgtdf wesnoth:master 5d617b9e4f27 / src/menu_events.cpp: Allow string identifers in :control http://git.io/vIqcn
20150605 15:40:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 15:41:55< Elvish_Hunter> gfgtdf: by the way, I'm in favor of moving [role] to Lua. I'm always in favor of moving C++ WML tags to Lua, if possible ;-)
20150605 15:43:53-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 15:48:25-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20150605 15:51:00< vultraz_iOS> Should we move all of them?
20150605 15:54:40< Elvish_Hunter> If reasonably possible, I think that all ActionWML tag should be. But this can be considered as a long term objective
20150605 15:54:49< Elvish_Hunter> *tags, not tag
20150605 15:58:44< vultraz_iOS> I might work on some of that once I return to my laptop
20150605 15:58:55< vultraz_iOS> Currently sitting in SFO so can't do anything :p
20150605 16:00:50< Elvish_Hunter> If I remember correctly, I should have somewhere a half-baked implementation of [set_variable] in Lua...
20150605 16:01:35< vultraz_iOS> It's that already Lua
20150605 16:02:04< vultraz_iOS> Stupid autocorrect seems to think Lua is actually either Luna or Lisa
20150605 16:02:10< vultraz_iOS> Isn't*
20150605 16:05:31< Elvish_Hunter> It isn't. It's in C++, line 1102 of src/game_events/action_wml.cpp
20150605 16:06:32-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 256 seconds]
20150605 16:10:20< Elvish_Hunter> Any opinion on my PR? https://github.com/wesnoth/wesnoth/pull/409
20150605 16:10:46< Elvish_Hunter> With this, I think that I finished my work on ImagePathWML :-)
20150605 16:16:07< gfgtdf> actually teh mian reason why i suggested iplementing [role] in lua is that its currently borken
20150605 16:19:36< Elvish_Hunter> I know. But converting a tag to Lua makes it easier to improve (when we want to add new options) and fix (like in this case)
20150605 16:26:35-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150605 16:26:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 255 seconds]
20150605 16:30:13-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150605 16:32:28-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 16:34:43-!- ancestral [~ancestral@71-34-1-180.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150605 16:41:18-!- gfgtdf [~chatzilla@x50abb053.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 38.0.5/20150525141253]]
20150605 16:42:36-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 16:46:36-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150605 16:54:12-!- stikonas_ is now known as stikonas
20150605 16:55:20-!- Elvish_Hunter [~irssi@wesnoth/developer/elvish-hunter] has quit [Quit: Ciao!]
20150605 17:09:22-!- [Relic] [~Relic]@2602:306:33a3:6d30:d9a7:d238:e1a0:a098] has joined #wesnoth-dev
20150605 17:39:39-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 17:56:51-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 18:35:50-!- irker090 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150605 18:56:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150605 19:05:06-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150605 19:19:05-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150605 19:25:36-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 19:55:04< shadowm> gfgtdf: IIRC fabi told me that side.controller could be a number, so it's probably that. It's just currently not saved by the editor.
20150605 19:57:51-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150605 19:58:59-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150605 19:59:05-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 20:05:25-!- hay207 [~haythamme@41.34.8.19] has joined #wesnoth-dev
20150605 20:22:12-!- hay207 [~haythamme@41.34.8.19] has quit [Quit: Leaving]
20150605 20:35:46-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 21:04:45-!- Kwandulin [~Miranda@p5B0085AA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150605 21:06:12< shadowm> Hm, gui2::tscrollbar_container::scroll_vertical_scrollbar() ought to work since its impementation is pretty much the same as the handle_key_*() methods. I guess the scrollbar's position is reset after pre_show() or something.
20150605 21:13:21-!- ancestral [~ancestral@252.sub-70-197-206.myvzw.com] has joined #wesnoth-dev
20150605 21:21:42-!- kex [~kex@31.11.67.182] has quit [Remote host closed the connection]
20150605 21:30:48-!- ancestral [~ancestral@252.sub-70-197-206.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20150605 21:38:39-!- OnkV [~quassel@85.124.51.57] has quit [Ping timeout: 264 seconds]
20150605 21:38:45-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 22:00:31-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150605 22:03:24-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds]
20150605 22:14:22-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 252 seconds]
20150605 22:22:52-!- gfgtdf [~chatzilla@x50abb053.dyn.telefonica.de] has joined #wesnoth-dev
20150605 22:23:30< gfgtdf> shadowm: hm yes i remember now. iirc this was a mp feature for mp campaigns.
20150605 22:24:35< gfgtdf> shadowm: i think i should replace it with a previous_save_id= attribute this is more flexible and doesnt interfere with sp.
20150605 22:25:12< gfgtdf> vultraz_iOS: you currently have a 1.12 branch ?
20150605 22:26:08< shadowm> I suspect his flight arrived already and he's no longer near a computer.
20150605 22:27:09< shadowm> gfgtdf: So, do you feel we can release 1.12.3 next Friday?
20150605 22:27:12< gfgtdf> shadowm: you know where he is or when he is back ?
20150605 22:27:26< shadowm> He said he'd be gone for around two weeks.
20150605 22:27:49< gfgtdf> shadowm: i think it would be nice if http://gna.org/bugs/?23635 is fixed before 1.12.3
20150605 22:27:52< shadowm> Maybe with very sporadic Internet access I imagine.
20150605 22:28:30< shadowm> Hm, I read that. I can't help with anything involving gamestate stuff or MP campaigns, unfortunately.
20150605 22:29:20< shadowm> Or is it just a bug with the campaign's WML?
20150605 22:29:32-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150605 22:29:34< gfgtdf> shadowm: it just fixing the save_id= and team_name=  attributes in LoW wml. (assuming that those  keys work just like te do in 1.13)
20150605 22:30:15< shadowm> Scenario 5 does have a save_id set for side 1 regardless of whether it's in MP mode or not.
20150605 22:31:01< shadowm> Side 2 appears to have no team name, though.
20150605 22:31:23< gfgtdf> shadowm: save_id= defaults to id=, so side 1 has a valid save_id, its just that the other multiplayer sides have neigher id= nor save_id=
20150605 22:32:01< gfgtdf> shadowm: no wait
20150605 22:32:10< gfgtdf> shadowm: side 1 has definitely save_id=Kalenz
20150605 22:32:19< gfgtdf> shadowm: https://github.com/wesnoth/wesnoth/blob/1.12/data/campaigns/Legend_of_Wesmere/scenarios/chapter2/05_The_Saurian_Treasury.cfg#L69
20150605 22:32:37< shadowm> Yes, that's what I said. :p
20150605 22:32:44< gfgtdf> shadowmah i missread
20150605 22:33:17< gfgtdf> shadowm: teh {PLAYABLE} macro sets team_name= for human sides
20150605 22:33:43< gfgtdf> shadowm: so basicly all human conrtolled sides should have {PLAYABLE}, and a save_id set
20150605 22:33:54< shadowm> Hm.
20150605 22:34:09< shadowm> Engine bug then. :(
20150605 22:34:17< gfgtdf> shadowm: why you think that ?
20150605 22:34:51< gfgtdf> shadowm: note that 'should' means 'should but are not' here
20150605 22:34:57-!- kex [~kex@31.11.67.182] has joined #wesnoth-dev
20150605 22:35:14< shadowm> Ah yes, side 2 and 3 don't have the macro there.
20150605 22:35:34< gfgtdf> shadowm: also no save_id
20150605 22:36:01< shadowm> The same situation affects scenario 6 where those become sides 6 and 7.
20150605 22:36:33< gfgtdf> shadowm: you know whether 1.13 and 1.12 have teh same fonts ?
20150605 22:36:35< shadowm> And 7 where they become 2 and 3 again.
20150605 22:36:54< shadowm> Except in scenario 7 they do have a team_name specified without the macro.
20150605 22:37:18< shadowm> And side 3 has a explicit save_id there but not side 2.
20150605 22:37:33< shadowm> gfgtdf: Yes, they do. Why?
20150605 22:37:39< gfgtdf> shadowm: it looks liek side 6 and 7 in scneario 6 dont even have contorller=human
20150605 22:38:15< gfgtdf> shadowm: becasue i aomstimes get some problems when i check out branches with different fontt foles becasue somehow windows keeps a lock on them
20150605 22:39:03< shadowm> It's probably our fault in a way.
20150605 22:39:32< gfgtdf> shadowm: you think we need to free the fontos in wesnoth at exit ?
20150605 22:39:39< gfgtdf> fonts*
20150605 22:40:01-!- kex [~kex@31.11.67.182] has quit [Ping timeout: 264 seconds]
20150605 22:40:13< shadowm> Yes, we currently use a more naïve approach where we register our font files with GDI on startup but never unregister them.
20150605 22:40:57< shadowm> GDI resources like this are owned by the Windows session process, which outlives Wesnoth.
20150605 22:41:24< shadowm> I can commit a fix later.
20150605 22:42:42< shadowm> Also, maybe you'd like to look into using git-new-workdir.cmd to maintain two checkouts of the same clone on disk? It's what vultraz and I use.
20150605 22:46:16< shadowm> Hm, or rather we should use AddFontResourceEx and mark the fonts as private to us.
20150605 22:53:59< gfgtdf> shadowm: i typed git-new-workdir in commandline and it says cannot find command.
20150605 22:55:41< shadowm> Of course, it's an external script.
20150605 22:56:04< gfgtdf> shadowm: h ok i'll try to donwload it
20150605 22:56:24< shadowm> It also requires messing with some system settings before using it on Windows 8 and 8.1 (not sure about 7).
20150605 22:56:51< shadowm> I don't remember the specifics because it's a one-time thing.
20150605 22:57:17< shadowm> Basically, it shares a single .git directory with two checkouts by using symbolic links.
20150605 22:57:53< shadowm> That means that the same repository exists in two directories at once, although looking at two different branches.
20150605 22:58:54< shadowm> So pulling from one means the objects fetched will be already available on the other (but you still need to pull from both to sync each to its respective upstream branch).
20150605 23:02:36< shadowm> gfgtdf: Hm, on closer inspection, we do call RemoveFontResource() when the font manager object is destroyed.
20150605 23:02:49< shadowm> Perhaps Wesnoth isn't exiting normally?
20150605 23:03:20< gfgtdf> shadowm: maybe it doesnt get freed when wesnoth crashes (which is likeley when you try to find teh casue for a crash)
20150605 23:04:30< shadowm> If it's a crash due to anything other than an exception being thrown then yes, the font manager destructor doesn't get called and the fonts are never removed.
20150605 23:05:06< shadowm> Same goes for killing the process with the task manager or from a debugger.
20150605 23:06:27< gfgtdf> shadowm: but marking as private would free it in that cases too ?
20150605 23:06:42< shadowm> According to MSDN, yes.
20150605 23:06:54< gfgtdf> shadowm: are there any reason why private is not teh default ?
20150605 23:07:22< shadowm> Who knows, I didn't write the Win32 API. :p
20150605 23:08:30< shadowm> As for why we don't do it I think it's just an oversight.
20150605 23:08:38-!- horrowind [~Icedove@2a02:810a:8b00:5298:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20150605 23:11:05< shadowm> Tested the patch, works fine.
20150605 23:13:12< gfgtdf> shadowm: which patch ?
20150605 23:14:01< gfgtdf> shadowm: AddFontResourceEx  or LoW wml ?
20150605 23:15:07< shadowm> The former.
20150605 23:16:52-!- irker682 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150605 23:16:52< irker682> wesnoth: Ignacio R. Morelle wesnoth:master e28cb897ad03 / changelog src/font.cpp: font: Mark our local fonts as private on Windows http://git.io/vIOWs
20150605 23:21:38< shadowm> And now I remembered that DejaVu Sans was upgraded in master but not in 1.12.
20150605 23:22:06< shadowm> master also has DejaVu Sans Mono, which 1.12 doesn't.
20150605 23:33:16-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20150605 23:35:16-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150605 23:50:45-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150605 23:54:36-!- cib0 [~cib@p508BCDE7.dip0.t-ipconnect.de] has quit [Ping timeout: 256 seconds]
--- Log closed Sat Jun 06 00:00:07 2015