--- Log opened Mon Oct 03 00:00:32 2016
20161003 00:20:06-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20161003 00:23:53-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 252 seconds]
20161003 00:23:53-!- wedge010 is now known as wedge009
20161003 00:25:32-!- Appleman1234 [~Appleman1@KD119104052245.au-net.ne.jp] has quit [Ping timeout: 252 seconds]
20161003 01:13:29-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20161003 01:13:29-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20161003 01:13:29-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20161003 01:17:58< tad_> Is the Lua console broken? It's taking me about 10-15 seconds just to see a character echoed
20161003 01:19:38< vultraz> tad_: seems to work alright for me
20161003 01:20:07< vultraz> (tested at titlescreen)
20161003 01:20:09< tad_> I just pulled master and re-built and it's very, very slow.
20161003 01:20:34< tad_> I'm inside new campaign/HttT and using debug/inspect as soon as able
20161003 01:21:35< tad_> I thought maybe I'd broken it so I rebuilt without my GL_Lua changes and no help.
20161003 01:25:31-!- Appleman1234 [~Appleman1@KD119104057150.au-net.ne.jp] has joined #wesnoth-dev
20161003 01:26:36< tad_> Actually, it's taking N*5 seconds .. N being the number of characters in the command text box.
20161003 01:26:45< vultraz> odd
20161003 01:26:51< vultraz> this problem, I do not have
20161003 01:27:46< tad_> I guess I can roll back a couple days and see if it's something recent.
20161003 01:28:07< vultraz> for the record
20161003 01:28:42-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 264 seconds]
20161003 01:29:06< vultraz> at certain times, under underteminate circumstances, gui2 will perform very badly
20161003 01:29:09-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20161003 01:29:12< vultraz> slowing to a crawl
20161003 01:29:29< vultraz> it's happened to me once or twice but under no reproducible circumstances or on any regular basis
20161003 01:30:06< tad_> So should I reboot? make clean/make? It's pretty consistent for me.
20161003 01:30:23< vultraz> a restart of wesnoth usually works
20161003 01:31:01< tad_> btdt several times. switched from GL_Lua to master and no help. pulled -rebase'd master and no help.
20161003 01:31:53< vultraz> then I cannot say
20161003 01:32:50< tad_> I'm gonna do the clear-it-all thing on the off chance it's a cache problem or something like that. Reboot. IF that does not clear it, I s'pose I'll try rolling back a few days and if that does it if I can't see a culprit I guess it's bisect time.
20161003 01:36:38< tad_> btw .. I still have that problem with the splash screen being the wrong size for the window .. and not resizing when I resize the window. Takes a couple Normal/Zoom/Normal cycles to get is cleared up
20161003 01:37:27< celticminstrel> tad_: It has happened to me before.
20161003 01:37:57< celticminstrel> I think clearing the log usually helps; it's why I added that button, though it's a workaround rather than a fix,
20161003 01:38:08 * celticminstrel is talking about the Lua console in case that's unclear.
20161003 01:38:48< tad_> Well now it won't even start HttT .. just hanging there with a window which has been invalidated, never repainted, and showing whatever was frozen when it got the focus
20161003 01:38:55< celticminstrel> :|
20161003 01:39:53< tad_> Building terrian ... then frozen.
20161003 01:40:13< tad_> CPU 0% 
20161003 01:40:55< tad_> This was from a clear-it-all you're-a-new-install ...
20161003 01:42:03< vultraz> whaaaa
20161003 01:42:27< tad_> I thought .. "Hmm, maybe it's sound, that's been a problem before." So I went to preferences, turned off all 4 sounds checks .. "Close" .. freeze
20161003 01:43:34< vultraz> T's very troubling 
20161003 01:43:38< vultraz> T'is*
20161003 01:44:29< tad_> So restarted. Did one sounds at a time. clsoe was fine.  After 3rd sound was unchecked decided the Quit instead of uncheck the fourth .. freeze
20161003 01:44:30-!- gfgtdf [~chatzilla@x4e368fc1.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20161003 01:44:59< tad_> And the changes were no saved. sigh
20161003 01:45:00-!- gfgtdf [~chatzilla@x4e36af65.dyn.telefonica.de] has joined #wesnoth-dev
20161003 01:45:51< tad_> I'm gonna check everything is still cleared then reboot.
20161003 01:45:55-!- tad_ [~tadcarluc@173.217.65.103] has quit [Quit: Leaving]
20161003 01:49:43-!- tad_ [~tadcarluc@173.217.65.103] has joined #wesnoth-dev
20161003 01:50:06< tad_> Can't even run it and immediately press Quit .. freezes
20161003 01:50:56< vultraz> screwed up, something definitely is
20161003 01:54:14< tad_> bisect, I will, Master Yoda.
20161003 02:04:13< vultraz> odd it is to think that seen the original star wars, you must have, when release, it did.
20161003 02:05:59-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20161003 02:22:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161003 02:28:27-!- irker025 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20161003 02:28:27< irker025> wesnoth: Charles Dang wesnoth:master 74bf3428e94f / src/ (gui/dialogs/title_screen.cpp gui/dialogs/title_screen.hpp wesnoth.cpp): Title Screen: big code/formatting cleanup https://github.com/wesnoth/wesnoth/commit/74bf3428e94fa9e0f02b404358c1fdcd370674a7
20161003 02:28:55< vultraz> ^mostly just adding comments/newlines and shuffling code around to more logical places
20161003 02:49:16< shadowm>     str << "\n" << _("#(Invalid)");
20161003 02:49:19< shadowm> Hm...
20161003 02:49:54< shadowm> vultraz: The GUI2 Load Game dialog has this bit of GUI1 markup.
20161003 02:50:09< shadowm> It's about the first thing in tgame_load::evaluate_summary_string().
20161003 02:50:25 * shadowm doesn't even know why he had that file open.
20161003 02:50:28< vultraz> What would that be in gui1?
20161003 02:50:36< shadowm> That's GUI1 markup.
20161003 02:51:25-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 02:51:26< travis-ci> wesnoth/wesnoth#11272 (master - 74bf342 : Charles Dang): The build was broken.
20161003 02:51:26< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164533390
20161003 02:51:26-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 02:51:34< vultraz> Yes, but does it actually display anything or is it the equivalent of <span></span>
20161003 02:52:42< shadowm> There's no equivalent for the latter in GUI1 and it'd be completely pointless.
20161003 02:53:01< shadowm> # sets the color to #FF0000, so the Pango version of that would be:
20161003 02:53:22< shadowm>     str << "\n<span color='#f00'>" << _("(Invalid)") << "</span>";
20161003 02:54:02< shadowm> You can test this by creating a plain text file named whatever in the saves dir.
20161003 02:54:32< shadowm> Unless your clock is completely wrong, it'll be the first thing displayed and have the "(Invalid)" line on the left panel.
20161003 02:55:15< shadowm> How about hiding that obnoxious horizontal line below the summary when there aren't any sides to display?
20161003 02:57:26< vultraz> Obnoxious? :| 
20161003 02:57:35< vultraz> Also, that'd be slightly more effort 
20161003 02:57:36< shadowm> Yeah.
20161003 02:57:52< vultraz> Those lines have no ids.
20161003 02:57:58< shadowm> I already said that it's a suspiciously retro-looking artifact of your UI design style.
20161003 02:58:17< vultraz> crap :|
20161003 02:58:51< shadowm> <shadowm> blah blah blah people these days use colors and contrast instead of lines
20161003 02:58:59< shadowm> I guess you weren't paying attention. As usual.
20161003 02:59:11< vultraz> yes, I remember you saying something to that effect
20161003 03:00:30< shadowm> Also, is there a specific reason the Unknown leader icon isn't team-colored? https://dl.dropboxusercontent.com/u/21371130/screenshots/Screenshot_20161002_235956.png
20161003 03:00:48< shadowm> Is it something about the save or about the dialog's code?
20161003 03:01:34< shadowm>     warning scripting/lua: wesnoth.put_unit(x, y, unit) is deprecated. Use wesnoth.put_unit(unit, x, y) instead
20161003 03:01:54< shadowm> celticminstrel: I get this when starting the tutorial.
20161003 03:02:09< celticminstrel> Fun! At least it's trivial to fix.
20161003 03:09:20< shadowm> Oh goody, show_wesnothd_server_search() reinvents base_name() even though I'm pretty sure the latter predates it.
20161003 03:09:53< shadowm> directory_name(), even.
20161003 03:21:11-!- Polsaker [~Polsaker@wikimedia/botters.Polsaker] has quit [Remote host closed the connection]
20161003 03:23:23-!- Polsaker [~Polsaker@wikimedia/botters.Polsaker] has joined #wesnoth-dev
20161003 03:25:32-!- gfgtdf [~chatzilla@x4e36af65.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 49.0.1/20160922113459]]
20161003 03:33:14< irker025> wesnoth: Charles Dang wesnoth:master 8241b0d173b1 / src/gui/dialogs/title_screen.cpp: Fixup 74bf3428e94f https://github.com/wesnoth/wesnoth/commit/8241b0d173b198c119e20515aac51b5abd51c488
20161003 03:33:17< irker025> wesnoth: Charles Dang wesnoth:master 59c0675691ff / src/gui/dialogs/game_load.cpp: Game Load: fixed a piece of legacy GUI1 formatting https://github.com/wesnoth/wesnoth/commit/59c0675691ff6f81eff07a451d172f8c616e1d73
20161003 03:33:20< irker025> wesnoth: Charles Dang wesnoth:master 268f9beb1c4e / src/ (gui/dialogs/game_load.cpp save_index.cpp): Game Load: apply TC to unknown leader image https://github.com/wesnoth/wesnoth/commit/268f9beb1c4e7ed45d8e395d0430e23dc4ada22d
20161003 03:33:34< vultraz> blah, forgot a file again
20161003 03:33:45< shadowm>     sh: 1: ./wesnothd-debug: not found
20161003 03:34:06< irker025> wesnoth: Charles Dang wesnoth:master 051519b03858 / data/gui/window/game_load.cfg: Fixup 59c0675691ff https://github.com/wesnoth/wesnoth/commit/051519b0385836a6c91d19c3e2130c089846746f
20161003 03:34:20< shadowm> Okay 1) why is Wesnoth using system() or an unsafe exec function to launch wesnothd; 2) how the hell does it know about the suffix despite the caller for the file chooser having no idea about it?
20161003 03:36:12< shadowm> Oh yes, I remember now, it uses system().
20161003 03:37:09< shadowm> So filesystem::get_program_invocation() is in charge of providing the suffix for game_launcher::start_wesnothd() if the path to wesnothd isn't set in preferences already...
20161003 03:37:18< celticminstrel> ...wait, seriously.
20161003 03:37:29< shadowm> And it's pretty naïve about it.
20161003 03:37:40< celticminstrel> It uses system() to launch wesnothd?
20161003 03:37:43< celticminstrel> Where is this?
20161003 03:37:50< shadowm> #ifdef DEBUG    + "-debug"    #endif
20161003 03:37:59< shadowm> #ifdef _WIN32    + ".exe"    #endif
20161003 03:38:21< shadowm> The first one wouldn't be of any help with CMake builds.
20161003 03:38:31< celticminstrel> Or XCode builds.
20161003 03:39:21< shadowm>     std::string config = filesystem::get_user_config_dir() + "/lan_server.cfg";
20161003 03:39:29< shadowm>     std::string command = "\"" + wesnothd_program +"\" -c \"" + config + "\" -d -t 2 -T 5";
20161003 03:39:42< shadowm>     if (std::system(command.c_str()) == 0) { /* ... */ }
20161003 03:39:45 * shadowm ponders this.
20161003 03:39:46< celticminstrel> Why is it not using fork+exec? :|
20161003 03:39:52< celticminstrel> (Or the Windows equivalent.)
20161003 03:40:14< celticminstrel> (Potentially wrapped behind something like a Boost thing, I dunno.)
20161003 03:40:16< shadowm> If the path to the user config dir has magic shell characters in it that system() will produce unexpected results.
20161003 03:40:29< shadowm> Same with the path to the wesnothd binary.
20161003 03:40:45< shadowm> Both are user-overridable values with defaults that we do not control.
20161003 03:41:19< shadowm> Although the average case is that the first is /home/foobar/.config/wesnoth and the second is /usr/bin.
20161003 03:41:29< shadowm> Or /usr/local/bin.
20161003 03:41:52< shadowm> Still not nice.
20161003 03:42:10< celticminstrel> So are you planning to change to fork+exec for UNIX systems, and the Windows equivalent, or just leave it?
20161003 03:42:26< shadowm> I probably will leave it since it's tangential to what I'm doing.
20161003 03:42:35< shadowm> Note that the Windows version of the code does something even more ridiculous.
20161003 03:42:40< celticminstrel> I'll put it on my TODO list, then. No idea how long it'll be before I get to it.
20161003 03:43:35< celticminstrel> (In particular I would have to look up how to do it on Windows.)
20161003 03:43:43< celticminstrel> (I'm sure it's not hard, I just don't know how.)
20161003 03:43:49< shadowm> CreateProcess().
20161003 03:44:20< shadowm> Not to be confused with ShellExecute(), which is unsafe by design.
20161003 03:46:45< celticminstrel> Works great for opening URLs though. >_>
20161003 03:46:55< shadowm> Yep.
20161003 03:47:22< shadowm> We don't have a choice there short of hardcoding a political choice wrt web browser preference.
20161003 03:48:06< shadowm> Oh fun, apparently the Microsoft CRT's version of system() according to the documentation still passes its argument to cmd.exe.
20161003 03:48:30< celticminstrel> Oh, Wesnoth uses ShellExecute for that too?
20161003 03:48:31< shadowm> Oh right, that's why I've seen system("cls") thrown around in some circles. cls is a cmd internal. Makes sense.
20161003 03:48:40< shadowm> celticminstrel: Yes.
20161003 03:50:00< shadowm> I suspect there's some really convoluted and version-dependant mechanism to query URL handlers from the shell API instead of trusting file associations to be sane.
20161003 03:50:21< shadowm> Then again, if file associations aren't sane then there's no point in doing that either.
20161003 03:51:33< shadowm> Besides, at least back in the day, URL handlers were implemented using file type associations.
20161003 03:59:54< shadowm> Do we have a way to sort a container full of strings, in a case-insenstive fashion?
20161003 04:00:21< celticminstrel> std should be able to do that.
20161003 04:00:30< shadowm> std is a namespace.
20161003 04:00:46< celticminstrel> Yes I know. What I mean is I'm pretty sure there's something in std that can do it.
20161003 04:00:57< shadowm> I know there are generic sort algorithms but I don't know what comparison function to pass to them.
20161003 04:01:53< shadowm> Okay, hang on a moment. The answer appears to be right in front of me.
20161003 04:02:32< shadowm> boost::algorithm::ilexicographical_compare, probably?
20161003 04:03:00< shadowm> Doesn't say whether it's locale-dependant but hopefully it is.
20161003 04:03:03< celticminstrel> Seems like there are lots of options. >_>
20161003 04:03:23< celticminstrel> If you absolutely want to ensure it's locale-dependent, apparently boost::locale has a comparator class.
20161003 04:03:59< celticminstrel> http://stackoverflow.com/a/10330109/1502810
20161003 04:04:01< shadowm> I probably shouldn't use that until Boost.Locale becomes mandatory across all configuration sets?
20161003 04:04:07< celticminstrel> Maybe...
20161003 04:04:12 * celticminstrel pokes vultraz 
20161003 04:04:17< vultraz> what?
20161003 04:04:21< celticminstrel> (Did he ever answer that question?)
20161003 04:04:46< celticminstrel> vultraz: Theoretically Wesnoth still supports libintl.
20161003 04:05:01< vultraz> what?
20161003 04:05:09< celticminstrel> As a substitute for Boost.Locale.
20161003 04:05:38< vultraz> and?
20161003 04:05:46< tad_> Didn't someone say yesterday they wanting libintl removed?
20161003 04:05:49< celticminstrel> shadowm asked you something about that yesterday, something like, "Is there any reason this hasn't been removed yet?"
20161003 04:06:14< vultraz> I said I knew of none
20161003 04:07:03< celticminstrel> So should it be removed?
20161003 04:07:20< vultraz> also wouldn't a plain std::sort work for shadowm's problem
20161003 04:07:23< celticminstrel> Should we ask localization people?
20161003 04:07:34< celticminstrel> vultraz: His question is how to get the comparator function.
20161003 04:07:51< vultraz> eh?
20161003 04:08:04< shadowm> vultraz: No, because it defaults to using an abstraction for operator<, and operator< for basic_string is case-sensitive.
20161003 04:08:24< celticminstrel> The only way a plain std::sort would work without a comparator is if you used basic_string<char, case_insensitive_traits>, where "case_insensitive_trats" is a class that you define.
20161003 04:09:03< celticminstrel> I'm pretty sure "operator< for basic_string is case-sensitive" is not a general truth. For string, yes, but not necessarily basic_string.
20161003 04:09:04< tad_> Check Boost.Locale.Collation?
20161003 04:09:06< shadowm> *for string is case-sensitive
20161003 04:09:11< vultraz> oh I see
20161003 04:09:38< shadowm> Yes, you are right, but I probably don't want to have a case-insensitive clone of string just for this one task.
20161003 04:10:07-!- travis-ci [~travis-ci@ec2-54-197-173-136.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 04:10:08< travis-ci> wesnoth/wesnoth#11273 (master - 268f9be : Charles Dang): The build was fixed.
20161003 04:10:08< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164539697
20161003 04:10:08-!- travis-ci [~travis-ci@ec2-54-197-173-136.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 04:10:20< celticminstrel> Probably not, and in any case it doesn't actually solve the problem, as you still need to find the comparator somewhere to use in that custom case_insensitive_traits class.
20161003 04:10:25< tad_> I think there's an example in boost.locale.collation making the comparator you need
20161003 04:10:32< vultraz> Aren't we using boost::locale for all builds anyway?
20161003 04:10:51< shadowm> Sigh.
20161003 04:11:04< shadowm> Our builds, perhaps. Not everybody's, yet.
20161003 04:11:37< celticminstrel> You'd have to ask all the Linux packagers for that question.
20161003 04:11:46< shadowm> Mailing list.
20161003 04:12:21< shadowm> I _think_ not forcing our OS X packagers to rebuild crap was also an argument in keeping the libintl-based implementation around.
20161003 04:12:25< celticminstrel> Right, post to the mailing list about "Wesnoth is considering dropping support for libintl as a substitute for Boost.Locale, any problems".
20161003 04:12:44< shadowm> But as with the older Ubuntu LTS argument, that's become kind of moot after the compiler mess for C++11.
20161003 04:12:52< celticminstrel> shadowm: But our OSX packager builds with Boost.Locale as far as I know. Unless he uses a modified XCode project.
20161003 04:13:13< shadowm> Then maybe I'm remembering wrong.
20161003 04:13:48< celticminstrel> Well, you could be remembering something quite a bit older.
20161003 04:13:50< celticminstrel> I dunno.
20161003 04:13:53< shadowm> I'm quite certain OpenPandora was an argument as well, but again, we don't support that anymore. And the latest Ubuntu LTS surely carries a decent Boost version.
20161003 04:13:58 * tad_ chucles. Google for libintl vs boost and our forums are #2 on the list ...
20161003 04:14:08< shadowm> Yes, I was talking in terms of what was discussed at the time version 1.11.19 was not yet released.
20161003 04:14:17< tad_> https://forums.wesnoth.org/viewtopic.php?f=10&t=41196
20161003 04:14:45< shadowm> The latest Ubuntu LTS at the time (14.04?) was also not the same one as now, which is 16.04.
20161003 04:15:10< shadowm> tad_: I wouldn't trust Google's results to be unbiased in that regard.
20161003 04:15:11< tad_> Ubuntu LTS is 16.04 IIRC
20161003 04:15:42< shadowm> Unless you were using a private window without cookies and without logging into any Google accounts you may have.
20161003 04:15:47< tad_> Nah. But it's interesting. Android was the issue for boost vs libintl in 2014 for 1.11
20161003 04:22:02< tad_> OK. I can't get Wesnoth 1.12.6 to run either. So it must be something about my system.
20161003 04:23:52< vultraz> D:
20161003 04:24:41< vultraz> tad_: for the record, this bug exists for the sluggish lua console https://gna.org/bugs/?24897
20161003 04:24:55< vultraz> and sigfurd just reported it's possibly worse recently
20161003 04:25:25< celticminstrel> So we need to find out if libintl is still needed for Android.
20161003 04:25:39< celticminstrel> For the record, I'm pretty sure it won't build with libintl at the moment.
20161003 04:25:46-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20161003 04:27:52< tad_> I suppose it would help to go into gdb and see where it's hanging up.
20161003 04:29:49< shadowm> /home/shadowm/src/wesnoth/src/gettext.cpp:110:37: error: no matching function for call to ‘lg::logger::dont_log(lg::log_domain& (&)())’
20161003 04:30:02< shadowm> Yeah, can't build with libintl at the moment.
20161003 04:30:42< tad_> gettext.cpp is for libintl. try gettext_boost.cpp .. the files are mutually exclusive
20161003 04:30:55< shadowm> Yes, I know this and that was on purpose.
20161003 04:31:03 * tad_ nods.
20161003 04:31:15< shadowm> I set ENABLE_LIBINTL to ON in CMake just out of curiosity to see if it'd build.
20161003 04:32:10< shadowm> Hm.
20161003 04:36:37-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 04:36:38< travis-ci> wesnoth/wesnoth#11274 (master - 051519b : Charles Dang): The build was fixed.
20161003 04:36:38< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164539809
20161003 04:36:38-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 04:36:44< shadowm> But yeah, the point is that either vultraz as RM or whoever will do the actual work ought to ask/warn that Boost.Locale is going to take over as the only option.
20161003 04:37:33 * tad_ chuckles.
20161003 04:37:35< celticminstrel> The libintl code could be fixed if really necessary, I guess.
20161003 04:38:04< shadowm> AFAICT it's just some divergence after some log API change, so yeah.
20161003 04:38:48< celticminstrel> Yeah. There'd probably also be a link failure due to the compare function only being in gettext_boost.cpp.
20161003 04:49:18-!- tad_ [~tadcarluc@173.217.65.103] has quit [Quit: Leaving]
20161003 05:12:00-!- Kwandulin [~Miranda@p200300760F2C716F0CB89332A4D3D176.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161003 05:31:15< shadowm> Okay, so enlighten me.
20161003 05:31:26< shadowm> I have:   namespace bl = boost::locale;  std::sort(entries.begin(), entries.end(), bl::comparator<char, bl::collator_base::secondary>{});
20161003 05:31:37< shadowm> This throws std::bad_cast.
20161003 05:32:32< shadowm> I think this means that I'm using an incompatible std::locale for the task?
20161003 05:36:17-!- atarocch [~atarocch@37.177.14.199] has joined #wesnoth-dev
20161003 05:39:48< shadowm> ... Okay, I guess we never use std::locale::global() ?
20161003 05:41:19< shadowm> But then how do I get the correct locale? It seems that's internal information mantained by gettext_boost.cpp.
20161003 05:41:28< shadowm> *maintained
20161003 05:44:42< shadowm> I guess I have to add a case-insensitive version of translation::compare()...
20161003 05:45:06< celticminstrel> shadowm: Yeah, so the only way to get it would be to expose it in gettext.hpp. I think I was mentioning this earlier too. (During the discussion of locale stuff in Lua.)
20161003 05:45:27< celticminstrel> Or the thing you just said.
20161003 05:45:33< shadowm> But this raises the question then, what's the global locale for us?
20161003 05:45:59< celticminstrel> My guess would be that it never touches the global locale, meaning it's probably the C locale.
20161003 05:46:02< shadowm> Is it whatever was specified by the operating environment, or do we override it?
20161003 05:47:09< shadowm> "a copy of std::locale::classic if no call to std::locale::global has been made." Okay, so C.
20161003 05:48:59< shadowm> So how would one go about obtaining the environment-defined locale? Just a lot of manual getenv() calls to check for LANG and the LC_* variables?
20161003 05:54:47< shadowm> Oh, I see, the empty string-identified locale.
20161003 05:58:00-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has joined #wesnoth-dev
20161003 06:02:40-!- tad_ [~tadcarluc@173.217.65.103] has joined #wesnoth-dev
20161003 06:04:35< tad_> vultraz, The problem with Wesnoth hanging for me was that it did not like not having a sound system. Once I configured my Arch for Advanced Linux Sound (ALSA) and got sound working, Wesnoth started working. I am getting a LOT of sound underrun errors on the console but it's working
20161003 06:05:24< shadowm> ... You... didn't have a sound stack configured before?
20161003 06:05:43< tad_> Correct
20161003 06:06:09< shadowm> DIY distributions. :\
20161003 06:06:15< tad_> Now I see ALSA lib pcm.c:8251:(snd_pcm_recover) underrun occurred
20161003 06:06:24< shadowm> Yes, ignore that, I've been getting it ever since 2008.
20161003 06:06:52< tad_> The point is that Wesnoth should gracefully handle sounds issues like not having one at all.
20161003 06:07:21< shadowm> I wouldn't even know how to test that without seriously crippling my whole system.
20161003 06:07:58< vultraz> Why should it be up to us to handle a lack of a sound system?
20161003 06:08:22< tad_> Well, it seems the error comes back and Wesnoth hangs.
20161003 06:08:28< shadowm> I assume I'd need to move away libasound and watch as half of all other applications start misbehaving.
20161003 06:08:46< celticminstrel> vultraz: Robustness?
20161003 06:08:59< tad_> Don't know why it just started today.
20161003 06:09:53< tad_> I am make'ing against current master and will see if getting sound config'd (so I can disable it in Preferences) fixes the problem with debug/inspect/lua-console too
20161003 06:10:19< shadowm> There is --nosound.
20161003 06:11:34< tad_> Well, there is probably also an error and we're re-trying in a hard loop instead of giving up and assuming --nosound was set. I can play with it now that I know the cause.
20161003 06:12:27-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161003 06:12:28< shadowm> For reference, using SDL_AUDIODRIVER=null gives me a functioning Wesnoth with sound automatically disabled.
20161003 06:12:30< tad_> Oh. And no help for the Lua console .. still taking 5 seconds per character on the line .. at about 5 or 6 characters it's over a minute to echo a character .. "slow" does not come close to the issue.
20161003 06:12:59< shadowm> But I don't know if SDL 2.0 falls back to that in the absence of everything else.
20161003 06:13:04< tad_> shadowm, So you're recommending we build it like that to avoid issues of systems with no sound cards?
20161003 06:13:09 * tad_ chuckles.
20161003 06:13:12< shadowm> That's not a build option.
20161003 06:13:55< vultraz> why would you even have a system with no sound card :|
20161003 06:14:09< tad_> Well, I'll see, tomorrow, if I can re-create the hanging issue and see if I can find where it's hanging.
20161003 06:15:11< tad_> vultraz, Because some of us think tech should be mute! I cut the vocal cords on my car, too. If I want to be nagged at while driving, I have a wife who does the job far better than and computer!
20161003 06:15:13< shadowm> vultraz: No sound *stack*, that doesn't necessarily mean there isn't a sound controller device.
20161003 06:15:38< vultraz> shadowm: he said no sound *cards*.
20161003 06:15:43< tad_> I had hardware. I didn't install the drivers.
20161003 06:15:58< shadowm> vultraz: He didn't.
20161003 06:16:05< vultraz> tad_: me thinks you don't use Siri.. :P
20161003 06:16:05< tad_> I'm on a VM so Linux sounds is sorta jerky anyway.
20161003 06:16:22< tad_> Siri? Isn't that a star?
20161003 06:16:35< vultraz> shadowm: "[17:13:03]	tad_	shadowm, So you're recommending we build it like that to avoid issues of systems with no sound cards?" <- *cards*
20161003 06:16:55< shadowm> vultraz: Don't make me quote the messages preceding that.
20161003 06:17:03< tad_> Of course not. And Cortana is disabled. And every device which does a damned thing if I say "Hello, Google" gets an immediate down-grade
20161003 06:17:13< vultraz> heh
20161003 06:17:33< tad_> vultraz, No card. Card but no drivers. All are errors we need to gracefully handle.
20161003 06:17:47< vultraz> I can't say I agree with you re mute tech but whatever.
20161003 06:18:04< vultraz> I think tech talking/listening to you is very helpful
20161003 06:18:33< shadowm> A deaf person installing a DIY distribution like Arch would probably have no use in installing ALSA userland other than satisfying dependencies.
20161003 06:18:38< tad_> I think it's annoying and I really don't like the idea that every sound in the room goes to Google.
20161003 06:20:01< tad_> Or Apple. Or the NSA. Or the local mafia don who is getting a feed from a Russian hacker who broke into the Siri voice stream ...
20161003 06:20:15< irker025> wesnoth: Charles Dang wesnoth:master 38680bde3b58 / data/gui/window/mp_create_game/ (default_resolution.cfg small_resolution.cfg): MP Create: remove vertical spacer lines https://github.com/wesnoth/wesnoth/commit/38680bde3b58ffbfd0a6760c1941176fe84e971b
20161003 06:20:20< vultraz> .... uh huh
20161003 06:20:26< vultraz> yes, that totally happens :|
20161003 06:20:50< tad_> The NSA taking a live feed from Google?
20161003 06:21:07< tad_> Or the mafia don puchasing the feed from the hacker?
20161003 06:21:12 * tad_ chuckle.s
20161003 06:21:15< vultraz> latter
20161003 06:22:01< celticminstrel> The star is Sirius, not Siri. >_>
20161003 06:22:04 * celticminstrel flees.
20161003 06:22:29< vultraz> ^
20161003 06:22:40< shadowm> So yeah, great. I need to change the local gettext API.
20161003 06:22:55< shadowm> And rebuild everything that uses it. Which is pretty much literally all the things. Yay.
20161003 06:23:03< vultraz> "// Set up eras menu_button" overzealous find/replace in action :P
20161003 06:25:14-!- atarocch [~atarocch@37.177.14.199] has quit [Ping timeout: 252 seconds]
20161003 06:26:00< celticminstrel> How so?
20161003 06:26:22< vultraz> it used to say combobox
20161003 06:26:30< celticminstrel> Exactly.
20161003 06:30:46-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20161003 06:37:50< shadowm> Yay, C++.
20161003 06:38:12< tad_> ?
20161003 06:41:44 * tad_ makes a note to self: "Self, remember NEVER use Remote Desktop on the Host with sound set to play locally. It sounds terrible when the VM and RDC are both using the sound card for the same music track."
20161003 06:42:52< shadowm> I did a code thing I can't say in here for fear or giving a certain someone weird ideas. This resulted in two pagefuls of nigh-undecipherable errors. Then I tried fixing that and apparently didn't even save my changes.
20161003 06:43:08< shadowm> Hence "yay".
20161003 06:43:29< shadowm> Why am I even building -O0... :\
20161003 06:43:41< tad_> Ah. Yes, C++ errors can be rather obtuse.
20161003 06:44:48< shadowm> Oh my god my code works even though this is literally the first time I ever use the locale API or Boost.Locale.
20161003 06:46:50< shadowm> Should I handle exceptions from boost::locale::collator<...>::compare(), and if so how?
20161003 06:47:57< shadowm> (Is that even a thing that can happen for the file dialog alone without everything else that uses locales dying first?)
20161003 06:48:27< tad_> shadowm, For exceptions: if in doubt, do not catch.
20161003 06:51:06-!- tad_ [~tadcarluc@173.217.65.103] has quit [Remote host closed the connection]
20161003 06:51:29-!- atarocch [~atarocch@37.148.180.150.static.hosted.by.combell.com] has joined #wesnoth-dev
20161003 06:51:50-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20161003 06:53:33-!- Kwandulin [~Miranda@p200300760F2C716F0CB89332A4D3D176.dip0.t-ipconnect.de] has quit [Ping timeout: 272 seconds]
20161003 06:54:09-!- tad_ [~tadcarluc@173.217.65.103] has joined #wesnoth-dev
20161003 06:55:02-!- tad_ [~tadcarluc@173.217.65.103] has quit [Client Quit]
20161003 06:55:39< irker025> wesnoth: Charles Dang wesnoth:master dbbc0b741495 / data/gui/window/drop_down_list.cfg src/gui/dialogs/drop_down_list.cpp: Dropdown List: set the tooltip on the whole panel, instead of just individual el https://github.com/wesnoth/wesnoth/commit/dbbc0b7414959c73b817506abb24f15298f51e34
20161003 06:55:50< vultraz> much better, that works
20161003 06:55:59< shadowm> Well, I guess if the std::sort predicate throws then it's time to abandon ship anyway.
20161003 06:57:56-!- Kwandulin [~Miranda@p200300760F2C71B25D1C6D854520D490.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161003 07:00:16< irker025> wesnoth: Charles Dang wesnoth:master 99e114ab17c7 / src/gui/dialogs/multiplayer/mp_create_game.cpp: MP Create Game: display Era descriptions as menu item tooltips instead of in the https://github.com/wesnoth/wesnoth/commit/99e114ab17c7d8087a55f7ef801d7125a0fb0f2b
20161003 07:00:33< vultraz> ^ allows players to view descriptions of eras as they browse the list
20161003 07:00:38< vultraz> instead of having to click on each one
20161003 07:01:23< vultraz> not sure what to do about mods
20161003 07:01:31< vultraz> perhaps I'll follow a similar system
20161003 07:03:05-!- atarocch [~atarocch@37.148.180.150.static.hosted.by.combell.com] has quit [Ping timeout: 248 seconds]
20161003 07:09:39< shadowm> Time to get rid of the GUI1 filechooser API.
20161003 07:09:59< shadowm> Such an uninteresting and tedious task.
20161003 07:10:01-!- Kwandulin [~Miranda@p200300760F2C71B25D1C6D854520D490.dip0.t-ipconnect.de] has quit [Ping timeout: 272 seconds]
20161003 07:11:03-!- Kwandulin [~Miranda@p200300760F2C71B268495F32D03A1B35.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161003 07:15:17-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 252 seconds]
20161003 07:15:58-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20161003 07:28:37< shadowm> Hm.
20161003 07:29:37< shadowm> Maybe I should add an overload of gui2::tdialog::show() that takes an output reference to save the selected path to if the dialog succeeds?
20161003 07:30:50< vultraz> sounds... reasonable
20161003 07:30:59< shadowm> I could always make set_path(std::string&) have weird semantics instead but that doesn't seem like the best design choice to me.
20161003 07:37:28< celticminstrel> Overloading tdialog::show doesn't seem like a great idea...
20161003 07:37:30< shadowm> I could also make a variant on the execute() static function that actually takes a dialog r-value reference along with the path output reference.
20161003 07:37:58< shadowm> So gui2::tfile_dialog::execute(video, gui2::tfile_dialog{}.set_foo(bar), path); ...
20161003 07:38:05< shadowm> Okay no, on second thought that looks silly.
20161003 07:38:19< vultraz> that looks very silly
20161003 07:39:06< shadowm> Is `gui2::tfile_dialog::execute(video, {}.set_foo(bar), path);` legal? That'd look slightly less silly.
20161003 07:39:33< vultraz> somehow i doubt it
20161003 07:40:04< shadowm> Or I can just make it not be a class member, like the GUI2 message/error message dialog functions.
20161003 07:43:53< shadowm> Yeah, it's not legal.
20161003 07:44:30< vultraz> that'd be too simple
20161003 07:49:25< vultraz> celticminstrel: possible is it, do you think, to utilize the ttip class to make the player info dialog floating on-mouseover?
20161003 07:50:55< vultraz> celticminstrel: think, I do, to utilize a 'floating' behavior for different dialogs instead of generalizing a grid swap for toolitps, more sense makes.
20161003 07:51:30< shadowm> ...
20161003 07:51:54< shadowm> It's not SW day, is it?
20161003 07:53:10< vultraz> been watching much star wars content, I have.
20161003 07:55:34-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20161003 07:56:31< shadowm> This is disgusting.
20161003 07:56:49< vultraz> hm?
20161003 07:57:03< zookeeper> well that does make me want to check the logs immediately
20161003 07:57:15< shadowm> I should probably add an overwrite prompt to the dialog itself instead of keeping these hand-rolled loops that the editor uses.
20161003 07:57:54< shadowm> (editor::context_manager::save_map_as_dialog() and save_scenario_as_dialog())
20161003 07:58:37< shadowm> For that matter, now I remember that I need to bring back the "template" (read: file extension) option.
20161003 07:59:23< shadowm> ... What.
20161003 07:59:46< shadowm> The stuff going on with res and overwrite_res here is so utterly pointless.
20161003 07:59:51< celticminstrel> I have no idea if that's possible, but it sounds potentially possible. (Also, I only understood the first of your sentences.)
20161003 08:00:08< shadowm> What is?
20161003 08:01:24< celticminstrel> What vultraz pinged me about. Sorry. Not really paying attention.
20161003 08:01:26< vultraz> celticminstrel: in the past i suggested we allow the specification of arbitrary grids in tooltips.
20161003 08:01:48< vultraz> celticminstrel: i think perhaps making different dialogs use the ttip behavior is more reasonable
20161003 08:03:11< celticminstrel> I'm guessing any window can be used as a ttip, but if you're having different types of tooltips for different things, you might have some difficulty...
20161003 08:03:25< celticminstrel> Anyway, sleep. Sorry.
20161003 08:03:35-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20161003 08:06:13< zookeeper> vultraz, i hope you're not saying that the load game dialog size in sire's screenshot is as intended
20161003 08:07:07< zookeeper> namely the way the third file is cut off
20161003 08:08:14< shadowm> Oh great. Now that the file dialog follow symlinks, .wesnoth-1.13/data/add-ons/After_the Storm/../ is not .wesnoth-1.13/data/add-ons/.
20161003 08:21:10< shadowm> Ugh, not the in-game save map function tooooo
20161003 08:26:51< shadowm> Okay, I got all the filechooser call sites converted to use the GUI2 file dialog.
20161003 08:27:33< shadowm> Now I just need to readd that filename template thing... which will certainly require me to mess with the implementation of the GUI2 textbox widget. Oh no.
20161003 08:28:05-!- boucman_work [~boucman@gre92-5-82-237-199-7.fbx.proxad.net] has joined #wesnoth-dev
20161003 08:29:40< shadowm> ... Why does src/gui/dialogs/editor/custom_tod.hpp include filechooser.hpp?
20161003 08:30:00< zookeeper> vultraz, why did you assign the UtBS lua error to me? surely that belongs to celticminstrel or someone as usual
20161003 08:30:02< shadowm> As well as some other stuff that doesn't belong in the interface.
20161003 08:30:52< shadowm>     src/editor/controller/editor_controller.cpp:99:65: error: invalid use of incomplete type ‘struct font::floating_label_context’
20161003 08:30:58< shadowm> Yeah no way this was going to be that easy.
20161003 08:31:05< shadowm> Stupid chain inclusions.
20161003 08:48:04< shadowm> Updated PR #778 again with what I've got for now.
20161003 09:20:05-!- fabi [~fabi@176.5.20.202] has joined #wesnoth-dev
20161003 09:37:09-!- boucman_work [~boucman@gre92-5-82-237-199-7.fbx.proxad.net] has quit [Remote host closed the connection]
20161003 09:37:31-!- boucman_work [~boucman@fw-alt.idf.smile.fr] has joined #wesnoth-dev
20161003 09:43:24< shadowm> loonycyborg: Please remove the SDL_net dependency from the CMake recipe?
20161003 09:44:04-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has joined #wesnoth-dev
20161003 09:44:09-!- atarocch [~atarocch@37.118.8.113] has joined #wesnoth-dev
20161003 09:45:00< loonycyborg> Not sure I'm 100% qualified to do something like this, I'll give a chance to mordante to fix it first :P
20161003 09:45:35< shadowm> I thought mordante was still inactive?
20161003 09:45:57< shadowm> Besides, he can't even build Wesnoth right now.
20161003 09:46:14< shadowm> Fuck it I'll do it myself later. As usual.
20161003 09:46:39< vultraz> zookeeper: the last entry being cut off has been fixed in master
20161003 09:46:41< loonycyborg> would be a chance for him to do something
20161003 09:47:10< shadowm> If I were him the last thing I'd want for my welcome party is people asking me to fix things I haven't used in years.
20161003 09:47:19< Nobun> sorry for noob question: how can I 'git merge' in a way that I will apply the merge BOTH in my local repository AND in my upstream github repository?
20161003 09:47:32< shadowm> Especially while unable to properly test them.
20161003 09:47:59< Nobun> (need it becouse I am near to prepare the pull request for wmlxgettext, but I need to make order in my personal repository before)
20161003 09:48:25-!- atarocch [~atarocch@37.118.8.113] has quit [Ping timeout: 248 seconds]
20161003 09:48:30< shadowm> Nobun: You just have to push your branch to your repository after you're done with it.
20161003 09:48:48< loonycyborg> Nobun: not sure I understand. Seems just a matter of doing git merge locally and then pushing
20161003 09:49:00< Nobun> try to explain better
20161003 09:49:21< vultraz> [19:29:40]	shadowm	... Why does src/gui/dialogs/editor/custom_tod.hpp include filechooser.hpp?
20161003 09:49:25< Nobun> I will try to explain better my question, I mean
20161003 09:49:51< Nobun> In my repository (wmlxgettext-unoff) I have two branches both locally and in my github space
20161003 09:49:52< shadowm> vultraz: Is that you taking responsibility for that silliness? :)
20161003 09:49:56< vultraz> shadowm: if it's messy it's because it's old code that I wrote in 1.11 when I was a noooob.
20161003 09:50:19< shadowm> That was my first suspicion.
20161003 09:50:24< Nobun> I know how to merge the branch into master locally, but I am unsure how to update github to perform the same merge action
20161003 09:50:27< vultraz> shadowm: but it's because the dialog opens the file chooser dialog for the metadata paths and there was some filesystem function I used which you recommended
20161003 09:50:50< shadowm> Yeah, but still doesn't belong in the interface.
20161003 09:51:04< shadowm> If I call the ToD editor, the last thing I need to know is that it internally calls the filechooser dialog.
20161003 09:51:22< shadowm> (Don't fix it though. I'd rather not have to deal with merge conflicts.)
20161003 09:51:32< vultraz> well, how else would one select paths?
20161003 09:51:43< shadowm> Interface versus implementation.
20161003 09:51:56< shadowm> The header file (custom_tod.hpp) provides the interface I use to call the ToD editor.
20161003 09:52:26< shadowm> The cpp file (custom_tod.cpp) provides the implementation of the ToD editor. That's a blackbox for me, the caller. I don't have any idea what goes on in there and I don't need to know.
20161003 09:52:42< shadowm> If the ToD editor uses the file chooser internally, it uses the file chooser internally.
20161003 09:52:52< shadowm> That means it's an implementation detail. Something I don't need to know.
20161003 09:53:03< vultraz> Oh
20161003 09:53:05< shadowm> Id est, something that isn't part of the ToD editor's interface.
20161003 09:53:08< vultraz> I misunderstood what you were saying
20161003 09:53:50< vultraz> anyway, I need to redo that dialog anyway
20161003 09:54:08< shadowm> mordante: Are you using any non-default CMake options, or any external versions of libraries? I just built master(+private branch) on a clean jessie install.
20161003 09:54:26< shadowm> Successfully, that is.
20161003 09:55:42< shadowm> With GCC 4.9.
20161003 09:56:06< loonycyborg> Nobun: I'd expect git to do the right thing in such a situation.
20161003 09:56:20< loonycyborg> if you just git push
20161003 09:56:50< loonycyborg> at remote end it has enough info to determine that those two branches got merged
20161003 09:57:02< Nobun> ok thank, loonycyborg
20161003 09:57:10< loonycyborg> since same commits have same hashes both locally and remotely
20161003 09:57:17< shadowm> loonycyborg: SCons also seems to check for SDL_net.
20161003 09:57:25< loonycyborg> at master?
20161003 09:57:34< shadowm> Oh, okay, no, the function have_sdl_net() is unused.
20161003 09:57:58< shadowm> The raw_sockets option references the existence of SDL_net in its help text though.
20161003 09:58:06< Nobun> I'm doing the last two tests, but I'm confident they will be passed, so probably the pull request with wmlxgettext with lua plural forms support should be ready within today
20161003 09:58:54< loonycyborg> I think removing network_worker stuff obsoletes raw sockets
20161003 09:59:16< loonycyborg> I'll clean it up
20161003 10:01:23< irker025> wesnoth: Ignacio R. Morelle wesnoth:master 8b341411d6b5 / CMakeLists.txt cmake/FindSDL2_net.cmake src/CMakeLists.txt: cmake: Remove SDL_net-related checks and substitutions https://github.com/wesnoth/wesnoth/commit/8b341411d6b54e0c92aaeb22e57147279edd6133
20161003 10:05:25< matthiaskrgr> can I change the title of a ticket I made?
20161003 10:06:02< loonycyborg> depends on the kind of ticket
20161003 10:06:11< matthiaskrgr> https://gna.org/bugs/index.php?25144#history 
20161003 10:06:14< shadowm> If you are logged in I think you should be able to do that?
20161003 10:06:31< matthiaskrgr> I am 
20161003 10:06:35< matthiaskrgr> (logged in)
20161003 10:06:56< matthiaskrgr> but I can only vote, add comment or attach new files :(
20161003 10:07:07< shadowm> Maybe the option is only available to project members then. What would the new title be?
20161003 10:07:32< matthiaskrgr> just "crash when trying to start campaign" 
20161003 10:07:50< shadowm> Okay, done.
20161003 10:07:53< matthiaskrgr> thanks
20161003 10:08:23< irker025> wesnoth: Ignacio R. Morelle wesnoth:master 4fe822b91b56 / INSTALL: Drop mentions of SDL_net from documentati https://github.com/wesnoth/wesnoth/commit/4fe822b91b56dc9332a9bddf6c80a3cea29d9322
20161003 10:09:03< matthiaskrgr> it's funny that the tutorial still works o.O
20161003 10:09:10< irker025> wesnoth: Ignacio R. Morelle wesnoth:master ca5ef3263654 / projectfiles/VC12/README.md: Also drop mentions of SDL_net from documentation https://github.com/wesnoth/wesnoth/commit/ca5ef3263654b478157630f9ef314842ebbb8558
20161003 10:10:38< shadowm>     option(ENABLE_PANDORA "Add support for the OpenPandora by adding support for the resolution of 800x480 and switching to a special theme" OFF)
20161003 10:10:57< shadowm> Funny I've been saying for the past couple of days that this was dropped.
20161003 10:11:52< shadowm> Maybe people decided to leave the build config toggle as a keepsake...
20161003 10:13:38< vultraz> More like it was never really cleaned up.
20161003 10:14:03< vultraz> We kinda just said 'oh, we won't really be able to support this anymore', and then didn't bother to clean up code related to it
20161003 10:14:10< shadowm> Can you remind me whether there was an official decision to drop it in the first place? Maybe my memory made that up.
20161003 10:14:20< shadowm> I don't even remember if it was mine or yours.
20161003 10:14:46< shadowm> Okay, if "we" just said that then it's not official.
20161003 10:14:51< vultraz> I don't really remember. I remember a discussion whose conclusion was that Pandora support would be lost.
20161003 10:15:08< vultraz> This was agreed to be acceptable.
20161003 10:15:21< shadowm> Probably something wrt C++11.
20161003 10:15:30< shadowm> Or SDL 2 or a Boost library bump.
20161003 10:16:04< vultraz> Later, you decided dropping 800x480 was also acceptable since Pandora support had been dropped (whether officially or not) and supporting the new Preferences at that resolution would be difficult
20161003 10:16:22< shadowm> No, I'm fairly sure I didn't say just that.
20161003 10:17:01< shadowm> What I said I think was that we could just silently stop supporting 800x480 in an official fashion and see what happens.
20161003 10:17:38< shadowm> That means not going out of our way to support it, but not actively dropping any support code either.
20161003 10:18:00< vultraz> About 2 weeks ago I removed the remaining #ifdef PANDORA guarded code.
20161003 10:18:04< shadowm> And this was when I was in charge so that doesn't have much validity by itself anymore.
20161003 10:18:24< shadowm> src/sound_music_track.cpp:88:#if !defined(_WIN32) && !defined(__APPLE__) && !defined(PANDORA)
20161003 10:18:27< shadowm> src/game_preferences.cpp:983:#ifndef PANDORA
20161003 10:18:29< shadowm> Not the last one, it seems.
20161003 10:18:31< vultraz> (though there seems to be 2 places I miss which -
20161003 10:18:54< vultraz> I seem to have missed though
20161003 10:18:58< vultraz> those*
20161003 10:19:04< vultraz> will clean up
20161003 10:19:21< matthiaskrgr> in the main menu, is it intended there is a new tip shown every time the window is resized?
20161003 10:19:43< vultraz> 800x600 is also now the set minimum resolution and windows cannot be resized smaller
20161003 10:20:03< vultraz> matthiaskrgr: it's a side effect of closing/reopening the dialog
20161003 10:20:19< irker025> wesnoth: Ignacio R. Morelle wesnoth:master 2c5904ae0aa7 / CMakeLists.txt: cmake: Remove ENABLE_PANDORA option https://github.com/wesnoth/wesnoth/commit/2c5904ae0aa73e0879eb529bf20f3aa980ed0076
20161003 10:20:22< matthiaskrgr> oh
20161003 10:20:43< shadowm> vultraz: I notice you didn't remove the Pandora theme, though.
20161003 10:22:32< vultraz> Again, there was never an extensive cleanup of the Pandora stuff.
20161003 10:22:46< irker025> wesnoth: Charles Dang wesnoth:master e7258407ecf8 / src/ (game_preferences.cpp sound_music_track.cpp): Cleaned up remaining uses of PANDORA define https://github.com/wesnoth/wesnoth/commit/e7258407ecf851d69e1026133f7412262593c0f4
20161003 10:23:25< irker025> wesnoth: Ignacio R. Morelle wesnoth:master 7e1798f34236 / CMakeLists.txt src/CMakeLists.txt: cmake: Drop ENABLE_SDL_GPU option https://github.com/wesnoth/wesnoth/commit/7e1798f342364248f401ef9ce12adbdff2cca53d
20161003 10:23:31< vultraz> The impression I got at the time was that it was left in case support could be restored.
20161003 10:24:12< shadowm> vultraz, Aginor: ^ Consider this a preliminary step towards evicting SDL_gpu from the codebase. I'd not dare proceed further for now until those WIP branches that I believe involve the affected code are merged.
20161003 10:24:19< vultraz> But it's been too long and with the multitude of changes that have been applied, it's very unlikely that will happen.
20161003 10:24:54< shadowm> I certainly wouldn't want to be the one to drop a conflict bomb all over your work.
20161003 10:25:03< Aginor> shadowm: I approve
20161003 10:25:18< vultraz> I think Aginor intends to redo his branches anyway.
20161003 10:25:31< vultraz> Though I've been waiting for him to do the SDL_GPU removal
20161003 10:26:06< Aginor> yes, my branch got conflicted into oblivion and I lost momentum quickly followed by motivation
20161003 10:26:12< Aginor> so do not consider my branch
20161003 10:26:46< shadowm> Hm. Doesn't vultraz have his own SDL-heavy branch too?
20161003 10:27:12< shadowm> Something something surface cleanup?
20161003 10:27:19< vultraz> merged
20161003 10:27:42< shadowm> So can I get rid of the whole thing then?
20161003 10:27:52< vultraz> SDL_GPU? Sure
20161003 10:28:05< Aginor> yes
20161003 10:28:13< vultraz> It will make life easier for us.
20161003 10:28:21< Aginor> vultraz: how so?
20161003 10:28:40< Aginor> it's unmaintained code, it's simply less scrolling unless you use folding
20161003 10:29:05< vultraz> Avoids having to check my grep results for whether its in one of the GPU defines or not
20161003 10:29:05< shadowm> I always forget my editor can fold blocks.
20161003 10:29:38< shadowm>  I'll take a look later.
20161003 10:30:02< Aginor> I used coan last time I did that I think
20161003 10:30:16< Aginor> I'm trying to grep the history to find the commit
20161003 10:30:59< Aginor> right, wrong repo
20161003 10:31:03< Aginor> doesn't help with grepping
20161003 10:31:50< Aginor> 358bd1235f07d8d577bdb2bb0b4aafcff2e93093
20161003 10:31:55< Aginor> This removes all legacy SDL1.2 code. It was done by invoking: coan source --replace --no-transients -D"SDL_VERSION_ATLEAST(X, Y, Z)"=1 <file> on each file.
20161003 10:34:40< irker025> wesnoth: loonycyborg wesnoth:master 5a75a242c5ec / SConstruct: scons: remove obsolete raw_sockets option https://github.com/wesnoth/wesnoth/commit/5a75a242c5ecb9ba83e1092c419684845b6ce877
20161003 10:39:49< Aginor> ok, removed SDL_GPU now
20161003 10:39:54< Aginor> doing a build
20161003 10:40:03< Aginor> then I can push a branch and make a PR if it compiles
20161003 10:42:13< irker025> wesnoth: loonycyborg wesnoth:master fd1c94b2f851 / SConstruct: scons: clean up unused sdl_net helper https://github.com/wesnoth/wesnoth/commit/fd1c94b2f851ef9dce60ff413d60c822713f6a1b
20161003 10:42:32< Aginor> loonycyborg: are you cleaning up cmakelists too?
20161003 10:42:52< loonycyborg> nope, shadowm did it
20161003 10:42:56< Aginor> ah, yes
20161003 10:43:01< Aginor> I should've read backlog
20161003 10:43:03< Aginor> thanks :)
20161003 10:50:35-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20161003 10:55:31< irker025> wesnoth: Andreas Löf wesnoth:remove_sdl_gpu e497761d4e35 / src/ (83 files in 20 dirs): Remove SDL_GPU #idfefs and library https://github.com/wesnoth/wesnoth/commit/e497761d4e352f00506877f4ecf05389edf25323
20161003 10:57:55< Aginor> https://github.com/wesnoth/wesnoth/pull/808 for those interested
20161003 10:58:13< JyrkiVesterinen> "Showing  83 changed files  with 0 additions and 57,981 deletions."
20161003 10:58:17< Aginor> it touches a lot of things, so check with peple before merging or it'll be commit hell
20161003 10:58:17< JyrkiVesterinen> Wow. Excellent. :)
20161003 10:58:23< Aginor> s/commit/conflict/
20161003 10:58:42< Aginor> JyrkiVesterinen: mostly SDL_gpu I think
20161003 10:59:17< Aginor> and I'm not pressing the merge button on that one, I haven't been keeping up to date enough to know who's working on what
20161003 10:59:41< JyrkiVesterinen> Oh. We have the entire SDL_gpu library in the Wesnoth repository. I hadn't noticed.
20161003 10:59:56< Aginor> not for much longer
20161003 11:00:09< JyrkiVesterinen> Fortunately.
20161003 11:00:30< Aginor> and with that, I shall go to bed
20161003 11:00:38< Aginor> good luck :D
20161003 11:01:40-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20161003 11:03:56-!- Kwandulin [~Miranda@p200300760F2C71B268495F32D03A1B35.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20161003 11:11:32-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20161003 11:15:17-!- RatArmy [~RatArmy@om126211116091.13.openmobile.ne.jp] has joined #wesnoth-dev
20161003 11:19:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161003 11:34:08-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has joined #wesnoth-dev
20161003 11:42:30-!- Appleman1234 [~Appleman1@KD119104057150.au-net.ne.jp] has quit [Ping timeout: 264 seconds]
20161003 11:51:51-!- RatArmy [~RatArmy@om126211116091.13.openmobile.ne.jp] has quit [Ping timeout: 272 seconds]
20161003 11:52:01-!- Duthlet [~Duthlet@dslb-188-104-253-155.188.104.pools.vodafone-ip.de] has joined #wesnoth-dev
20161003 11:58:58-!- RatArmy [~RatArmy@om126211116091.13.openmobile.ne.jp] has joined #wesnoth-dev
20161003 12:04:35-!- louis94 [~~louis94@91.178.241.79] has joined #wesnoth-dev
20161003 12:07:18-!- RatArmy [~RatArmy@om126211116091.13.openmobile.ne.jp] has quit [Quit: Leaving]
20161003 12:15:56-!- fabi [~fabi@176.5.20.202] has quit [Read error: No route to host]
20161003 12:15:59< zookeeper> is there a way to undo a commit, as in take me back to exactly how things were before it? so, committed changes revert back to local changes, added files revert back to whatever-you-call-uncommitted-new-files, etc.
20161003 12:16:08-!- fabi [~fabi@176.5.20.202] has joined #wesnoth-dev
20161003 12:16:40-!- Appleman1234 [~Appleman1@KD106154018030.au-net.ne.jp] has joined #wesnoth-dev
20161003 12:18:15< zookeeper> i'm pretty sure none of the usual reset stuff would do that.
20161003 12:18:17-!- Bonobo [~Bonobo@2001:44b8:254:3200:2d2b:4929:31a:e4c] has quit [Quit: Leaving]
20161003 12:18:27< zookeeper> but maybe it would, i don't know.
20161003 12:19:57< zookeeper> ok, i think mixed reset did it right.
20161003 12:25:26-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161003 12:30:05< zookeeper> anyway, i figured i'd take the desert elf redesign through a PR for formality's sake, although i doubt there's much point to it: https://github.com/wesnoth/wesnoth/pull/810
20161003 12:36:22< matthiaskrgr> if you want to undo your last commit, you can do
20161003 12:36:27< matthiaskrgr> git reset HEAD~1
20161003 12:36:49< matthiaskrgr> (will still leave the modified files of the commit in repo, needs to rm/git checkout to go completely back) 
20161003 12:38:10-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20161003 12:39:19< mkdroid> if you want to modify your last commit, adding to HEAD and then git commit --ammend   might do
20161003 12:39:51< mkdroid> (srry,  network conb broke down)
20161003 12:42:01-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20161003 12:44:25-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 272 seconds]
20161003 12:49:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161003 12:49:50-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Remote host closed the connection]
20161003 12:52:01-!- louis94 [~~louis94@91.178.241.79] has quit [Ping timeout: 244 seconds]
20161003 13:07:28-!- Appleman1234_ [~Appleman1@KD106154000111.au-net.ne.jp] has joined #wesnoth-dev
20161003 13:08:56-!- Appleman1234 [~Appleman1@KD106154018030.au-net.ne.jp] has quit [Ping timeout: 252 seconds]
20161003 13:09:08-!- Appleman1234_ is now known as Appleman1234
20161003 13:11:22-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20161003 13:13:29-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20161003 13:21:23-!- tad_ [~tadcarluc@173.217.65.103] has joined #wesnoth-dev
20161003 13:25:37-!- Kwandulin [~Miranda@p200300760F2C71B245F439A00BDB884C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161003 13:27:26-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161003 13:53:33-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has joined #wesnoth-dev
20161003 13:56:01-!- irker025 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20161003 14:06:59< tad_> Well, now that I have sound working I can get Wesnoth up.
20161003 14:07:54< tad_> And I can see that the Lua console slowdown (sslloowwddoooowwww....) is only when coming from the gamestate inspector.
20161003 14:09:22< tad_> So, I was able to verify, finally, that I need to add setlocale to the whitelist for Lua "os" module.
20161003 14:13:53-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 14:13:54< travis-ci> wesnoth/wesnoth#11290 (master - fd1c94b : loonycyborg): The build has errored.
20161003 14:13:54< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164599256
20161003 14:13:54-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 14:37:53< vultraz> from a look at the modified file list in the remove SDL_GPU pr it looks like it won't conflict too much with any major WIP branch
20161003 14:38:07< vultraz> I'll confirm wit celticminstrel, though
20161003 14:38:39-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20161003 14:39:21-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 14:39:22< travis-ci> wesnoth/wesnoth#11291 (remove_sdl_gpu - e497761 : Andreas Löf): The build passed.
20161003 14:39:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164601442
20161003 14:39:22-!- travis-ci [~travis-ci@ec2-54-144-5-89.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 14:40:09-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 248 seconds]
20161003 14:57:29-!- Kwandulin [~Miranda@p200300760F2C71B245F439A00BDB884C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20161003 14:59:30< vultraz> tad_: that's what the bug report says
20161003 14:59:33< vultraz> interesting
20161003 15:00:40< vultraz> again, doesn't happen for me
20161003 15:01:11< tad_> vultraz, Yes well last night I was having a lot of problems all over the place and wanted to state I am back to 'known problems' state :)
20161003 15:01:55< tad_> I'm working on the Lua stuff for a while and will take a look at the lua console slowdown later.
20161003 15:03:30< vultraz> you have no objections if the sdl gpu branch is merged?
20161003 15:03:36< vultraz> remove sdl gpu*
20161003 15:07:05-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161003 15:12:50-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20161003 15:19:49-!- Kwandulin [~Miranda@p200300760F2C71B2FD6A36629A574668.dip0.t-ipconnect.de] has joined #wesnoth-dev
20161003 15:30:52< tad_> So let's say I have a C/Lua literal string. For example a unit type ID. First off, it's a utf-8 string. And, it should be only codepoints from 0x20 to 0x7E. In those cases, strcmp and strcoll are identical. So if I miss an os.setlocale it's probably not going to matter, anyway.
20161003 15:34:33-!- boucman_work [~boucman@fw-alt.idf.smile.fr] has quit [Ping timeout: 248 seconds]
20161003 15:36:49-!- boucman_work [~boucman@fw-alt.idf.smile.fr] has joined #wesnoth-dev
20161003 15:37:24-!- louis94 [~~louis94@91.178.241.79] has joined #wesnoth-dev
20161003 15:47:18-!- louis94 [~~louis94@91.178.241.79] has quit [Ping timeout: 264 seconds]
20161003 16:02:49-!- boucman_work [~boucman@fw-alt.idf.smile.fr] has quit [Ping timeout: 248 seconds]
20161003 16:02:58-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has joined #wesnoth-dev
20161003 16:04:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 16:04:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 16:04:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 16:08:35-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20161003 16:08:49< mordante> servus
20161003 16:10:54< mordante> shadowm, not sure I think I use standard flags, just updated an old checkout and did a make
20161003 16:11:06< mordante> loonycyborg, SDL_net is no longer required?
20161003 16:11:23< loonycyborg> yup
20161003 16:12:07< zookeeper> good grief, the orcish archer is positioned completely wrong...
20161003 16:12:50< mordante> loonycyborg, did you change it to boost:asio?
20161003 16:13:47< loonycyborg> yes
20161003 16:14:49< mordante> cool :-)
20161003 16:30:22-!- tad_ [~tadcarluc@173.217.65.103] has quit [Quit: Leaving]
20161003 16:33:19-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20161003 16:34:19-!- irker939 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20161003 16:34:19< irker939> wesnoth: ln-zookeeper wesnoth:master f350865b1d4e / data/core/images/units/orcs/ (28 files): Centered frames of Orcish Archer properly https://github.com/wesnoth/wesnoth/commit/f350865b1d4e4d66f49960146af448f50034e722
20161003 16:36:06< zookeeper> i wrote a gimp script to do that for me, so at least now i can quickly do it for a large number of images whenever necessary...
20161003 16:39:53-!- JyrkiVesterinen [~JyrkiVest@78-27-109-121.bb.dnainternet.fi] has joined #wesnoth-dev
20161003 16:40:42-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20161003 16:42:36-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has joined #wesnoth-dev
20161003 16:44:06-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20161003 16:48:31-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20161003 16:56:17-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20161003 16:58:14< irker939> wesnoth: ln-zookeeper wesnoth:master 1fbe90664b33 / data/campaigns/Heir_To_The_Throne/images/units/ (6 files): Centered frames of Sea Orc properly, and tweaked the shadow https://github.com/wesnoth/wesnoth/commit/1fbe90664b3363e0277b0b8ede5e2f70927ea5ef
20161003 16:59:49 * zookeeper facepalms
20161003 16:59:53< zookeeper> why do i need to fix everything myself?
20161003 17:00:03< zookeeper> start_time=-100 wtf is this
20161003 17:00:25< vultraz> why not just run that script on all mainline sprites
20161003 17:01:09-!- ancestral [~ancestral@75-168-189-115.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20161003 17:01:09< vultraz> also, why do leadership animations always seem to start with frame 2 :|
20161003 17:01:38-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20161003 17:01:52< zookeeper> do you think i wrote a gimp script implementing some kind of advanced AI which can detect how many pixels a given sprite needs to be shifted to put it in the right spot?
20161003 17:01:59< zookeeper> because i didn't
20161003 17:02:24< vultraz> I did :P
20161003 17:03:23< vultraz> why not make it run a Zealous Crop operation using max(w,h) and then use the centering tool to center it. Assuming you can control gimp tools from its scripting interface
20161003 17:03:59< zookeeper> i have no idea what you're saying or why you think the zealous crop tool would implement such an AI
20161003 17:04:32 * celticminstrel doesn't really like 99e114ab17c7d8087a55f7ef801d7125a0fb0f2b
20161003 17:05:21< vultraz> celticminstrel: whyyy not?
20161003 17:05:32< vultraz> celticminstrel: also, any objections to merging the sdl_gpu branch
20161003 17:05:44< celticminstrel> Not quite sure. It might be just because it's different.
20161003 17:06:42< irker939> wesnoth: ln-zookeeper wesnoth:master f894667eb1b1 / data/campaigns/Heir_To_The_Throne/units/Sea_Orc.cfg: Changed the weird Sea Orc attack animation timing https://github.com/wesnoth/wesnoth/commit/f894667eb1b1cb3b8fb99967aa2720c27ccdeae9
20161003 17:06:57< zookeeper> vultraz, this should all be your job since you chose to merge the PR prematurely
20161003 17:07:30< vultraz> nope, I know nothing of animation timings
20161003 17:07:36< zookeeper> exactly
20161003 17:07:44< celticminstrel> Then you shouldn't be merging animation PRs.
20161003 17:08:03< vultraz> zookeeper said nothing about there being a problem
20161003 17:08:22< celticminstrel> Did he say anything about whether it seemed ready?
20161003 17:08:40< vultraz> He said there were duplicate commits 
20161003 17:08:56-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20161003 17:10:15< zookeeper> well, whatever. clearly i was going to have to do it in any case.
20161003 17:10:47< vultraz> yup
20161003 17:12:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 17:15:54< celticminstrel> Sudden crab showed up. o.o
20161003 17:16:21< vultraz> so, should i merge the remove sdl_gpu branch?
20161003 17:16:25< celticminstrel> Oh hey, Nobun finished his plural support.
20161003 17:16:33< celticminstrel> Patience, young padawan.
20161003 17:16:34< vultraz> will it seriously conflict with anything you're doing?
20161003 17:16:56< zookeeper> i was just about to make a poor joke that you're lucky it was singular, but i dunno what to say now
20161003 17:18:13-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 17:24:53< mattsc> celticminstrel: yeah, I was surprised about crab commenting also
20161003 17:25:36< mattsc> So is anybody opposed to me merging PR 807 (speed improvements for AI with many guardians)?
20161003 17:29:23< irker939> wesnoth: ln-zookeeper wesnoth:master 85c12f40a5cd / data/campaigns/Heir_To_The_Throne/scenarios/04_The_Bay_of_Pearls.cfg: Removed Bugg's random trait and fixed his buggy transformation https://github.com/wesnoth/wesnoth/commit/85c12f40a5cd40e22448df200a41892711999b15
20161003 17:29:54< zookeeper> anyone happen to recall other instances in mainline campaigns where a unit changes/transforms/reveals into something else?
20161003 17:30:32< zookeeper> the cursed loyalists in liberty is one
20161003 17:31:16< zookeeper> jessene is another
20161003 17:33:27< celticminstrel> Why does PR 810 duplicate every scenario. :|
20161003 17:35:11< celticminstrel> Wait, why is log.cpp using C string manipulation? :|
20161003 17:35:22-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20161003 17:39:59< vultraz> zookeeper: besides thot? utbs garak?
20161003 17:40:35< vultraz> celticminstrel: likely because he wants the new scenarios to exist alongside the old ones temporarily
20161003 17:41:01< celticminstrel> Are the scenarios that different?
20161003 17:41:10< vultraz> no idea
20161003 17:41:12< vultraz> haven't looked
20161003 17:41:12< zookeeper> naturally, because of unit id's
20161003 17:41:14< vultraz> too much diff
20161003 17:41:20< vultraz> my browser cannot handle it
20161003 17:41:39< celticminstrel> I doubt it's your browser.
20161003 17:41:50< celticminstrel> It's probably github throttling the diff size.
20161003 17:42:04< celticminstrel> Unless you crashed or something.
20161003 17:42:12< vultraz> no, it lags horribly when i scroll on a huge duff\
20161003 17:42:15< vultraz> diff
20161003 17:42:59< mattsc> celticminstrel: you said that you wanted to look at PR 807.  Should I wait for that or assume that no news is good news?
20161003 17:43:09< celticminstrel> I'm looking now.
20161003 17:43:15< celticminstrel> Wait, this is 808.
20161003 17:43:20< celticminstrel> What was 807 again?
20161003 17:43:22< celticminstrel> Oh right.
20161003 17:43:29< celticminstrel> You're mattsc not vultraz. >_>
20161003 17:43:42< celticminstrel> Maybe I should change the highlight colour. Orange on purple is difficult to read.
20161003 17:43:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 17:43:48< mattsc> most of the time, at least  ;)
20161003 17:43:52< celticminstrel> (Or... change the nickname colour...)
20161003 17:44:17< celticminstrel> I looked at PR 807 already. Yesterday I think? Maybe the day before?
20161003 17:44:29< mattsc> Cool, thanks.
20161003 17:44:46< celticminstrel> I'm pretty sure I said nothing because I didn't see anything wrong or needs work or anything.
20161003 17:45:14< mattsc> I’ll wait a couple more hours to see if anybody else has comments and then merge it.
20161003 17:45:49-!- travis-ci [~travis-ci@ec2-23-20-167-49.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 17:45:50< travis-ci> wesnoth/wesnoth#11298 (master - f350865 : ln-zookeeper): The build passed.
20161003 17:45:50< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164681993
20161003 17:45:50-!- travis-ci [~travis-ci@ec2-23-20-167-49.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 17:47:06-!- fabi [~fabi@176.5.20.202] has quit [Remote host closed the connection]
20161003 17:48:01-!- ancestral [~ancestral@209.181.254.220] has joined #wesnoth-dev
20161003 17:48:02-!- fabi [~fabi@176.5.20.202] has joined #wesnoth-dev
20161003 17:49:56< celticminstrel> vultraz, 808 looks good to me.
20161003 17:50:37< celticminstrel> Other than the typo in the commit message. :P
20161003 17:51:26< vultraz> It lends character :P
20161003 17:51:36< celticminstrel> 9_9
20161003 17:52:30< celticminstrel> Minus 58k LoC.
20161003 17:52:55< vultraz> most of that is probably sdl_gpu itself
20161003 17:53:07< celticminstrel> Probably.
20161003 17:53:25 * celticminstrel assumes you mean "the actual source for SDL_GPU".
20161003 17:53:53< irker939> wesnoth: Andreas Löf wesnoth:master e497761d4e35 / src/ (83 files in 20 dirs): Remove SDL_GPU #idfefs and library https://github.com/wesnoth/wesnoth/commit/e497761d4e352f00506877f4ecf05389edf25323
20161003 17:53:55< irker939> wesnoth: Charles Dang wesnoth:master 0a4ff56d7601 / src/ (83 files in 20 dirs): Merge pull request #808 from wesnoth/remove_sdl_gpu https://github.com/wesnoth/wesnoth/commit/0a4ff56d7601c6f31af4c1fcbeeb2ed089f67ea6
20161003 17:56:09< vultraz> and that doesn't even conflict with shadowm's file dialog pr :D
20161003 17:56:21-!- fabi [~fabi@176.5.20.202] has quit [Remote host closed the connection]
20161003 17:56:32< vultraz> (well, he did introduce some conflicts by modifying cmkaelists, but that was earlier)
20161003 17:56:41< vultraz> cmakelist
20161003 17:56:43< celticminstrel> -849 LoC from removing the GUI1 filechooser.
20161003 17:58:32-!- ancestral [~ancestral@209.181.254.220] has quit [Quit: i go nstuf kthxbai]
20161003 17:59:09-!- fabi [~fabi@176.5.20.202] has joined #wesnoth-dev
20161003 17:59:13< vultraz> celticminstrel: thoughts on removing utils/shared_object.hpp?
20161003 17:59:25< celticminstrel> I don't know what it is.
20161003 18:03:05< vultraz> unused
20161003 18:03:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 18:03:52< vultraz> seems like some sort of primitive shared_ptr to me, but I don't really know.
20161003 18:10:46-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20161003 18:16:07-!- travis-ci [~travis-ci@ec2-23-20-167-49.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 18:16:08< travis-ci> wesnoth/wesnoth#11299 (master - 1fbe906 : ln-zookeeper): The build passed.
20161003 18:16:08< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164688345
20161003 18:16:08-!- travis-ci [~travis-ci@ec2-23-20-167-49.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 18:19:30< celticminstrel> Does SDL_NO_STDIO_REDIRECTION still do something?
20161003 18:19:41-!- Kwandulin [~Miranda@p200300760F2C71B2FD6A36629A574668.dip0.t-ipconnect.de] has quit [Quit: Kwandulin]
20161003 18:20:06< celticminstrel> ^-NO_ -ION
20161003 18:25:38-!- tad_ [~tadcarluc@173.217.65.103] has joined #wesnoth-dev
20161003 18:26:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20161003 18:45:40-!- travis-ci [~travis-ci@ec2-54-205-232-20.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 18:45:41< travis-ci> wesnoth/wesnoth#11300 (master - f894667 : ln-zookeeper): The build passed.
20161003 18:45:41< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164691870
20161003 18:45:41-!- travis-ci [~travis-ci@ec2-54-205-232-20.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 18:49:13-!- JyrkiVesterinen [~JyrkiVest@78-27-109-121.bb.dnainternet.fi] has quit [Quit: .]
20161003 18:49:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161003 19:05:57-!- travis-ci [~travis-ci@ec2-54-205-232-20.compute-1.amazonaws.com] has joined #wesnoth-dev
20161003 19:05:58< travis-ci> wesnoth/wesnoth#11301 (master - 85c12f4 : ln-zookeeper): The build passed.
20161003 19:05:58< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/164699362
20161003 19:05:58-!- travis-ci [~travis-ci@ec2-54-205-232-20.compute-1.amazonaws.com] has left #wesnoth-dev []
20161003 19:07:38-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Ping timeout: 244 seconds]
20161003 19:08:00-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20161003 19:10:18-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has joined #wesnoth-dev
20161003 19:10:50< Nobun> celticminstrel: as you can see I added a pull request with wmlxgettext with plural support
20161003 19:11:11< Nobun> In the pull request message I reported all the tests I did... if something is not clear, feel free to ask.
20161003 19:11:41< Nobun> Obliouvsly, feel free to make any additional tests that can be required
20161003 19:14:17-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20161003 19:17:58-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20161003 19:17:59-!- wedge010 is now known as wedge009
20161003 19:17:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20161003 19:18:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161003 19:22:42< celticminstrel> Nobun: Yeah, I saw.
20161003 19:24:03< Nobun> I only marked it, since I hope I could be enogh clear on explaining what tests I did :P
20161003 19:24:24< Nobun> and, if not. I will try to do my best to explain myself better xD
20161003 19:25:18-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20161003 19:39:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 19:40:31-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 272 seconds]
20161003 19:44:23-!- gfgtdf [~chatzilla@x4e36af65.dyn.telefonica.de] has joined #wesnoth-dev
20161003 19:59:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 19:59:51-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20161003 20:03:08< Aginor> celticminstrel: what typo?
20161003 20:03:17< Aginor> celticminstrel: I reject your baseless accusations!
20161003 20:07:40< celticminstrel> Haha
20161003 20:13:57-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20161003 20:23:20< vultraz> celticminstrel: so do you think I should make mods in mp create also use tooltips for descriptions?
20161003 20:23:36< vultraz> tbh, having the descriptions show to the left is rather confusing
20161003 20:27:31< celticminstrel> I think I prefer to see the descriptions on left, but whatever.
20161003 20:37:08-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20161003 20:37:08-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20161003 20:54:21< vultraz> wait, there's a reevaluate_best_size = key in [resolution]?
20161003 20:54:23< vultraz> huh
20161003 20:54:30-!- irker939 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20161003 20:58:46< vultraz> hm
20161003 20:58:54< vultraz> we don't have a gradient draw function, do we?
20161003 21:04:31-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 21:06:03-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 21:06:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: No route to host]
20161003 21:06:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 21:09:03-!- mjs-de [~mjs-de@x4e30e829.dyn.telefonica.de] has joined #wesnoth-dev
20161003 21:10:52< shadowm> mordante: Maybe you should try deleting the CMake build dir and starting a new one from scratch then. I've often had it break on me after some amount of changes to the recipe/source tree, not just with Wesnoth even.
20161003 21:11:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 21:13:26< Nobun> yes, I confirm it is a nice tip
20161003 21:14:08< Nobun> whenever somethings changes in source code file structure, the best thing is to make another build dir from scratch
20161003 21:14:17-!- Nobun [~nobun@host219-23-dynamic.50-82-r.retail.telecomitalia.it] has quit [Quit: Salve a tutti]
20161003 21:15:21< shadowm> Oh cool, I guess Aginor beat me to dropping SDL_gpu.
20161003 21:15:43< vultraz> celticminstrel: anything on the flg dialog?
20161003 21:15:59< celticminstrel> Sorry, still haven't looked at it.
20161003 21:17:47< vultraz> :(
20161003 21:19:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 21:19:56< Aginor> shadowm: it wasn't much work for me since I had the old commands I used when dropping SDL1.2 so I figured I'd just do it last night
20161003 21:20:04< Aginor> shadowm: so sorry if I stole work from you
20161003 21:20:34< vultraz> #FirstWorldProblems 
20161003 21:23:35 * shadowm stares at vultraz blankly.
20161003 21:24:11< vultraz> I'm making a joke that it's a FWP that one might complain that someone else made less work for them :P
20161003 21:25:00< shadowm> Okay? I don't even live in the "first world" so it doesn't really work for me.
20161003 21:25:58< vultraz> Doesn't mean you can't have FWP
20161003 21:26:14< vultraz> s
20161003 21:30:02< Aginor> vultraz: that's inappropriate
20161003 21:30:15< vultraz> how so?
20161003 21:32:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20161003 21:34:07-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20161003 21:54:18-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20161003 21:56:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 265 seconds]
20161003 22:03:23-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20161003 22:04:02-!- mjs-de [~mjs-de@x4e30e829.dyn.telefonica.de] has quit [Remote host closed the connection]
20161003 22:08:49-!- mkdr0id [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20161003 22:22:33-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 248 seconds]
20161003 22:49:30-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20161003 22:53:47< shadowm> Oh, so GUI2 textboxes simply do not have a selection... :\
20161003 22:56:03< celticminstrel> Huh?
20161003 22:56:09< celticminstrel> What are you talking about?
20161003 22:56:45< shadowm> Oh, they do.
20161003 22:56:54< shadowm> It's just that FOR SOME REASON it's not highlighted.
20161003 22:57:17< celticminstrel> Ah, it's not? I might've vaguely noticed that in the back of my mind at some point...
20161003 22:58:17< shadowm> It is highlighted in 1.12, for the record.
20161003 22:59:13< shadowm> Looks like someone dropped a relevant part of the canvas WML.
20161003 22:59:51< shadowm> vultraz, of course. In commit 67cf95a5af3caefb3123da806126282997c1121d .
20161003 23:02:29-!- irker795 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20161003 23:02:29< irker795> wesnoth: Ignacio R. Morelle wesnoth:master b1fe37b90b86 / changelog data/gui/widget/text_box_default.cfg: gui2/ttext_box: Restore selection highlight lost in version 1.13.3 https://github.com/wesnoth/wesnoth/commit/b1fe37b90b868608bc75067235fb8bd8c91146dd
20161003 23:07:07< irker795> wesnoth: Ignacio R. Morelle wesnoth:master 5375ad838fd8 / data/gui/widget/text_box_default.cfg: gui2/ttext_box: Move selection highlight to the right place https://github.com/wesnoth/wesnoth/commit/5375ad838fd8b6c32df0795e71fe5a24da3ba059
20161003 23:08:10< shadowm> So this somehow causes the initial state to include a 1 pixel-wide selection highlight box.
20161003 23:09:22-!- ancestral [~ancestral@215.sub-174-219-5.myvzw.com] has joined #wesnoth-dev
20161003 23:11:15-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20161003 23:16:59< shadowm> Did anyone change the GUI2 canvas code recently?
20161003 23:19:06< shadowm> I guess the answer is: yes, a lot.
20161003 23:23:50< shadowm> It seems that commit e993e2820b26e244ee0fa1f8df0f5c16d2bad763 might've caused this. SDL_RenderFillRect() has an effect when the rectangle width is zero which I guess we don't want.
20161003 23:27:08< shadowm> Aginor: Do you know if it's correct for SDL_RenderFillRect() to paint a 1 pixel wide rectangle if width is zero?
20161003 23:34:24-!- ancestral [~ancestral@215.sub-174-219-5.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20161003 23:34:29< shadowm> My gut feeling is "no", but the documentation doesn't say anything at all.
20161003 23:35:13-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20161003 23:36:33< shadowm> Nor does anyone else on the Internet as far as I can tell.
20161003 23:43:38-!- gfgtdf [~chatzilla@x4e36af65.dyn.telefonica.de] has quit [Ping timeout: 252 seconds]
--- Log closed Tue Oct 04 00:00:45 2016