--- Log opened Tue Aug 30 00:00:06 2016
20160830 00:03:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 00:11:55< irker082> wesnoth: Charles Dang wesnoth:master 7baf9587ad9f / / (3 files in 2 dirs): MP Lobby: implement all replay view options https://github.com/wesnoth/wesnoth/commit/7baf9587ad9fb5d501816b8c2a202e1bcf38de71
20160830 00:22:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 00:22:32< vultraz> ok, what else was never added to the new lobby...
20160830 00:22:43< vultraz> besides addon dependency downloads..
20160830 00:23:30< celmin> Is it possible to combine quick and blinded?
20160830 00:23:51-!- louis94 [~~louis94@91.178.242.69] has quit [Ping timeout: 264 seconds]
20160830 00:24:01< vultraz> not as currently presented
20160830 00:25:42< vultraz> remind me at some point to combine those preferences into an enum
20160830 00:28:39< vultraz> hm
20160830 00:28:42< vultraz> what is std::unary_function
20160830 00:29:28-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160830 00:29:55< celmin> Probably a C++98 precursor of std::function.
20160830 00:30:07< vultraz> "unary_function is deprecated in C++11."
20160830 00:30:09< vultraz> oh deer
20160830 00:30:44< celmin> There are no deer here.
20160830 00:30:48-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 265 seconds]
20160830 00:30:49-!- wedge010 is now known as wedge009
20160830 00:30:51< celmin> Only pigeons.
20160830 00:31:09< vultraz> usage here https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/lobby/data.hpp#L190
20160830 00:31:23< vultraz> trying to figure out why the hell there are all these filter classes P_P
20160830 00:31:33< celmin> It's probably fine to just remove the inheritanec.
20160830 00:31:35< celmin> ^inheritance
20160830 00:31:59< celmin> The inheritance from unary_function, that is. Not the inheritance from game_filter_base.
20160830 00:32:10< vultraz> ah, there's one more use of unary_function in the source code
20160830 00:32:36< vultraz> https://github.com/wesnoth/wesnoth/blob/master/src/pathutils.hpp#L24
20160830 00:33:03< celmin> I imagine that too can just drop the inheritance.
20160830 00:33:32< vultraz> what even does that class do :|
20160830 00:33:50< celmin> The one inheriting from unary_function?
20160830 00:33:52< vultraz> it defines a virtual operator 
20160830 00:34:05< vultraz> the xy_pref class, yes
20160830 00:34:07< vultraz> pred
20160830 00:34:13< celmin> Pure virtual, to be precise.
20160830 00:34:49< vultraz> terrain_filter inherits from it but doesn't do anything with it..
20160830 00:35:12< vultraz> and its only use is as a type of the last argument of get_tiles_radius
20160830 00:36:14< celmin> What that means is that terrain filters can be passed to get_tiles_radius.
20160830 00:36:50< vultraz> then why not just of type terrain_filter?
20160830 00:37:25< celmin> That would exclude other similar things from being passed.
20160830 00:37:41< celmin> Is terrain_filter the only class that inherits from it?
20160830 00:37:59< vultraz> yes
20160830 00:38:25< vultraz> and the only time any get_tiles_radius is called with that last argument, is is with objects of type terrain_filter
20160830 00:38:30< vultraz> it is*
20160830 00:39:09< vultraz> and nothing else inherits from xy_pred
20160830 00:39:14< vultraz> meaning nothing else CAN be passed
20160830 00:39:16< vultraz> :|
20160830 00:40:45< vultraz> would you say the inheritance can be dropped and the argument changed?
20160830 00:40:57< celmin> I don't think it's worth it, honestly.
20160830 00:41:37< vultraz> well i'm going to be touching the code anyway
20160830 00:41:39< vultraz> might as well
20160830 00:41:50< celmin> I really don't think it's worth it.
20160830 00:42:10< vultraz> how come?
20160830 00:42:19< celmin> By leaving it that way we can more easily add new xy_preds in the future. For example, one that calls a Lua or WFL function.
20160830 00:42:30< vultraz> i see
20160830 00:42:45< vultraz> so you just recommend deleting both cases of inheritance from unary_function?
20160830 00:44:38< celmin> Yeah.
20160830 00:45:04< vultraz> alright
20160830 00:45:27< celmin> I wonder where that function is called from though...
20160830 00:45:58< vultraz> next i need to figure out this ridiculous lobby games filter system
20160830 00:47:46< irker082> wesnoth: Charles Dang wesnoth:master 9b782931bbd4 / src/ (gui/dialogs/lobby/data.hpp pathutils.hpp): Removed cases of inheritance from the deprecated std::unary_function https://github.com/wesnoth/wesnoth/commit/9b782931bbd441469a76e8b355d59bbe74cce005
20160830 00:48:46< vultraz> what...
20160830 00:50:29< vultraz> so when you filter..
20160830 00:50:32< vultraz> it calls add_game_filter
20160830 00:50:45< vultraz> construct a new pointer to game_filter_value...
20160830 00:50:54< vultraz> why??
20160830 00:52:04< vultraz> i see
20160830 00:52:14< vultraz> these classes are used to get shown/not shown filters 
20160830 00:53:31< vultraz> why can't it keep a vector of bools 
20160830 00:53:47< vultraz> iterate over the list of games
20160830 00:54:04< celmin> Don't forget that vector<bool> is not a vector.
20160830 00:54:04< vultraz> and just check some data
20160830 00:54:35< vultraz> oh 
20160830 00:54:40< vultraz> it DOES
20160830 00:54:46< vultraz> it DOES keep a vector of bools :|
20160830 00:56:07< vultraz> ok, I guess the hard part was specifying the elements to check
20160830 00:57:48< vultraz> ok
20160830 00:58:07< vultraz> it seems to be constructing objects holding the member and value
20160830 00:59:47< vultraz> i can't help but think there must be a better way...
20160830 01:01:45< vultraz> ok I have 
20160830 01:01:48< vultraz> an idea
20160830 01:03:29< vultraz> hmm
20160830 01:03:35< vultraz> celmin: what does this do? game_info::*member_
20160830 01:10:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 01:15:14< celmin> Looks like a pointer-to-member?
20160830 01:15:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 258 seconds]
20160830 01:15:34< vultraz> perhaps
20160830 01:15:39< vultraz> don't need that, though
20160830 01:15:48< vultraz> i think i figured out a good way to do this
20160830 01:15:59< celmin> With a type in front it would declare a variable "member_" of type "pointer to member of game_info of type".
20160830 01:16:10< celmin> Where is this?
20160830 01:16:38< vultraz> https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/lobby/data.hpp#L245
20160830 01:17:13< celmin> So I guess the idea is that the template filters on a member of game_info.
20160830 01:17:28< celmin> Are you trying to fix something that isn't broken?
20160830 01:17:28< vultraz> yes
20160830 01:17:33< vultraz> simplify it
20160830 01:17:44-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 01:17:45< travis-ci> wesnoth/wesnoth#10606 (master - 7baf958 : Charles Dang): The build has errored.
20160830 01:17:45< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156090538
20160830 01:17:45-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 01:17:45< vultraz> i think I came up with a way
20160830 01:17:48< celmin> Well, if it truly does simplify it, I suppose that's okay, but...
20160830 01:17:50< vultraz> will push to my fork soon
20160830 01:20:30< celmin> I'm trying to wrap my head around the matrix placement again.
20160830 01:21:17< celmin> I tried an alternate way (minimizing area/perimeter), but something must've been very wrong with it, because the grid didn't even appear.
20160830 01:21:18< vultraz> no pun intended? :P
20160830 01:21:25< celmin> Huh?
20160830 01:21:40< celmin> I don't see this pun.
20160830 01:21:44< vultraz> wrap around?
20160830 01:21:46< vultraz> blah
20160830 01:21:49< vultraz> bad pun
20160830 01:21:52< celmin> Oh, I get it.
20160830 01:22:06< celmin> The placement is about wrapping cells, after all.
20160830 01:23:09< celmin> It's not obvious how to decide which size is "best".
20160830 01:25:00< vultraz> hmm
20160830 01:25:06< vultraz> slight snag..
20160830 01:25:17< celmin> I suspect that extra space below it on lower resolutions occurs because it calculated its best height but then expanded its width so that it didn't need so much height.
20160830 01:25:24< vultraz> ah, nvm
20160830 01:25:51< celmin> I could say the best size is "all in one row", but that seems weird.
20160830 01:26:31< celmin> I could say the best size is a fixed number of rows (eg 2, or 3), but that seems too arbitrary.
20160830 01:26:40< vultraz> hm
20160830 01:26:46< vultraz> now i haven't seen this error before..
20160830 01:26:48< celmin> …maybe I can update the size at placement time?
20160830 01:26:49< vultraz> C:\TDM-GCC-32\lib\gcc\mingw32\5.1.0\include\c++\ext\new_allocator.h|63|error: forming pointer to reference type 'std::function<bool(const game_info&)>&'|
20160830 01:27:06< celmin> You can't have a pointer to a reference.
20160830 01:27:13< celmin> A reference to a pointer is fine.
20160830 01:27:32< celmin> What's the template instantiation that produced that error?
20160830 01:27:52< celmin> ie, look for "In instantiation of" or similar lines in the error log.
20160830 01:28:39< vultraz> eh, already fixed
20160830 01:28:44< celmin> Yay.
20160830 01:29:32< vultraz> for some reason it didn't like me keeping the vector members by reference
20160830 01:29:48< celmin> You can't have a vector of references.
20160830 01:29:59< celmin> Or a vector of const anything.
20160830 01:30:05< vultraz> celmin: anyway, here's the general idea: https://github.com/Vultraz/wesnoth/commit/e41a0b00cc611a63641039a642f96a1267fe2f92
20160830 01:30:12< celmin> Or a vector of anything that does not have a copy constructor, if I recall correctly.
20160830 01:30:50< vultraz> i have spotted a flaw, though..
20160830 01:31:17< vultraz> the result of the functions are cumulative.. 
20160830 01:31:21< vultraz> aren't*
20160830 01:31:23< celmin> Your loop at line 335 looks wrong.
20160830 01:31:23< vultraz> results*
20160830 01:31:29< celmin> Shouldn't it be show = show && …?
20160830 01:31:38< celmin> Although you could short-circuit.
20160830 01:32:08< celmin> So I guess, if((show = show && f())) break;
20160830 01:32:31< celmin> (The redundant layer of parentheses there silences a "did you mean equals" compiler warning.)
20160830 01:32:54< celmin> I guess you don't need the && in that case, just show = f() would suffice.
20160830 01:33:03< celmin> Hmm, I'm mixed up.
20160830 01:33:30< celmin> You'd need to break if false, not if true, assuming you're going for "all must match". If you want "any must match", then if((show = f()) break; would work.
20160830 01:33:57< vultraz> all must match, I think
20160830 01:34:01< vultraz> which makes sense
20160830 01:34:05< celmin> Then just put ! in front
20160830 01:34:10< vultraz> if i filter on a and b i want games with a and b
20160830 01:34:12< celmin> if(!(show = f())) break;
20160830 01:34:34< celmin> Hmm, "and stack" huh.
20160830 01:34:52< vultraz> hm?
20160830 01:35:39< celmin> I always find it weird when I see ==true or ==false
20160830 01:35:48< vultraz> anyway, do you agree this method is a lot simpler and easier to read than the old one?
20160830 01:36:09< celmin> It seems simpler, but I have this suspicion that it may be less flexible.
20160830 01:36:52< vultraz> how so?
20160830 01:38:39< celmin> The old version seems to be set up to create arbitrarily complex boolean expressions.
20160830 01:38:51< celmin> (Although there's no game_filter_or_stack, it would be trivial to define one.)
20160830 01:41:01< celmin> Then again, I may not be clear on just what you're actually removing.
20160830 01:42:05< vultraz> the filter classes
20160830 01:42:44< celmin> What about game_filter_general_string_part?
20160830 01:43:32< vultraz> moving its match() implementation to a function in game_info.
20160830 01:43:49< vultraz> so I can just do info.match_string_filter(string)
20160830 01:44:03< celmin> This definitely sounds less flexible, then.
20160830 01:44:19< vultraz> not really?
20160830 01:44:49< vultraz> the values it filtered on (map_info and name) were hardcoded
20160830 01:45:09< celmin> The lobby didn't make use of the full power that the filter system supported, yes.
20160830 01:45:23< celmin> And the OR stack was never even implemented.
20160830 01:45:31< celmin> There was NOT filter though.
20160830 01:45:36< celmin> Which seems like it was never used?
20160830 01:45:39< celmin> ^was a
20160830 01:46:34< vultraz> there's an Invert option
20160830 01:46:37< vultraz> in the lobby
20160830 01:47:01< celmin> Ah.
20160830 01:47:15< celmin> So it was used for that?
20160830 01:47:20< vultraz> it just flips the shown result
20160830 01:47:41< vultraz> all the results are considered, then, if invert was selected, flipped
20160830 01:47:49< vultraz> er
20160830 01:47:52< vultraz> the result is flipped
20160830 01:48:32< vultraz> there's no OR settings
20160830 01:48:38< vultraz> there're
20160830 01:49:21< celmin> So, most of the filter options are actually only a toggle, right?
20160830 01:49:40< vultraz> yes
20160830 01:50:17< celmin> And there's just the four ways to filter.
20160830 01:50:26< celmin> And the GUI1 filtering was basically the same?
20160830 01:51:38< vultraz> yes
20160830 01:51:44< vultraz> but without No Ignored
20160830 01:52:01< celmin> Can you think of any reason you would want OR filters?
20160830 01:52:10< vultraz> not really
20160830 01:52:18< vultraz> since you can just deselect options you don't want
20160830 01:52:23< vultraz> final commit: https://github.com/Vultraz/wesnoth/commit/31a359bac7321fc3c11eb2daf5d78bac3c3801b3
20160830 01:52:25< celmin> I suppose the lobby doesn't really need arbitrary boolean expressions.
20160830 01:52:47< celmin> Unless they're there for scripting purposes or something.
20160830 01:53:14< vultraz> either way, these conditionals are *much* easier to understand
20160830 01:53:16< vultraz> and add
20160830 01:53:51< celmin> I think that's an exaggeration, but whatever.
20160830 01:54:42< vultraz> well, what would you rather see: "return info.vacant_slots > 0" or "new game_filter_value<size_t, &game_info::vacant_slots, std::greater<size_t> >(0))"
20160830 01:54:43< vultraz> :P
20160830 01:55:03< celmin> They're both fairly clear to me, though I guess not equally so.
20160830 01:55:23< vultraz> plus
20160830 01:55:31< vultraz> the filter can still be as complex as we want
20160830 01:55:35< vultraz> it just has to return a bool
20160830 01:55:51< celmin> Oh, you took my braceless if-statement literally, huh.
20160830 01:56:11< vultraz> that's something the old system could not do
20160830 01:56:14< vultraz> as far as I can tell
20160830 01:56:18< vultraz> it could only compare one value
20160830 01:56:29< vultraz> with my system, we could add a complex filter that compares many values
20160830 01:56:49< celmin> Nah, you could easily have defined your own complex filter in the same way as the string matching one.
20160830 01:57:15< celmin> ie, a new subclass
20160830 01:57:21< vultraz> booo
20160830 01:57:35< vultraz> this is easier for future coders :P
20160830 01:57:37< celmin> It's about as much work as defining a new lambda.
20160830 01:57:45< celmin> Only very slightly more.
20160830 01:58:00< celmin> So, those filter classes were not used anywhere else?
20160830 01:58:33< vultraz> no
20160830 02:00:26< vultraz> HMMMM
20160830 02:00:35< vultraz> the filtering works, but not in the way I'd expect..
20160830 02:00:35< celmin> ALL CAPS
20160830 02:00:42< celmin> How so?
20160830 02:00:57< vultraz> if no filtering options are selected, it displays no games
20160830 02:01:34< celmin> "bool show = false;"
20160830 02:01:42< celmin> That's probably the culprit
20160830 02:02:47< vultraz> ah, yes
20160830 02:04:10< vultraz> seems to work alright
20160830 02:04:29< vultraz> the number of games displayed label doesn't update, though
20160830 02:04:31< vultraz> different issue
20160830 02:04:57< irker082> wesnoth: Charles Dang wesnoth:master 95a2fb3a46f3 / src/gui/dialogs/lobby/ (data.cpp data.hpp info.cpp info.hpp lobby.cpp): MP Lobby: refactored games filtering https://github.com/wesnoth/wesnoth/commit/95a2fb3a46f3ef7460ff37766827cf0fcfcb38e7
20160830 02:05:16-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 02:05:17< travis-ci> wesnoth/wesnoth#10607 (master - 9b78293 : Charles Dang): The build has errored.
20160830 02:05:17< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156095310
20160830 02:05:17-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 02:13:15-!- Shiki [~Shiki@141.39.226.227] has quit [Remote host closed the connection]
20160830 02:22:42< vultraz> hm
20160830 02:23:02< vultraz> this is odd
20160830 02:23:08< vultraz> the 'shown' label, that is
20160830 02:23:13< vultraz> num shown*
20160830 02:28:21< vultraz> well now i got it to display the opposite of what's visible
20160830 02:28:22< vultraz> heh
20160830 02:58:58-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 02:58:59< travis-ci> wesnoth/wesnoth#10608 (master - 95a2fb3 : Charles Dang): The build has errored.
20160830 02:58:59< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156104227
20160830 02:58:59-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 03:03:58-!- JyrkiVesterinen [~JyrkiVest@87-100-161-181.bb.dnainternet.fi] has joined #wesnoth-dev
20160830 03:57:09-!- JyrkiVesterinen [~JyrkiVest@87-100-161-181.bb.dnainternet.fi] has quit [Quit: .]
20160830 04:27:33-!- midzer_ [~quassel@p5B2960C2.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160830 04:30:03-!- midzer [~quassel@p5B2969B8.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds]
20160830 04:46:10-!- mjs-de [~mjs-de@x4db66424.dyn.telefonica.de] has joined #wesnoth-dev
20160830 04:51:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 04:55:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160830 05:05:07-!- irker082 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160830 05:15:11-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160830 05:25:58-!- jswensen [~jswensen@cpe-98-145-147-33.natnow.res.rr.com] has joined #wesnoth-dev
20160830 05:26:04-!- mjs-de [~mjs-de@x4db66424.dyn.telefonica.de] has quit [Remote host closed the connection]
20160830 05:32:10< jswensen> FcConfigGetCurrent() is failing on the iOS real device, but not on the simulator. I’m not sure why.
20160830 05:32:12-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20160830 05:32:42< jswensen> This is in font.cpp
20160830 05:35:21< jswensen> in font::manager::update_font_path it is trying to add the font config path to the current fontconfig configuration, but FcConfigGetCurrent() fails. Is there some fontconfig initialization step that needs to take place first?
20160830 05:42:10-!- jswensen [~jswensen@cpe-98-145-147-33.natnow.res.rr.com] has quit [Quit: jswensen]
20160830 05:44:29-!- Kwandulin [~Miranda@p200300760F4241BF5CB29195B8D86938.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160830 05:46:33< celmin> Well, I think I figured out why the matrix placement policy put them all in one column.
20160830 05:48:46< celmin> I was comparing a coordinate with a size.
20160830 05:49:03< celmin> They're both (somewhat inconveniently) stored as tpoint.
20160830 06:06:07< wedge009> mattsc: Are there perhaps fixed-units in UMC? I can't recall any standard unit which can't move.
20160830 06:08:54< celmin> I think there was something in TSG.
20160830 06:13:29-!- TC02 [~quassel@venus.arosser.com] has quit [Quit: No Ping reply in 180 seconds.]
20160830 06:14:47-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160830 06:15:50< wedge009> Which scenario? o.O
20160830 06:16:16< celmin> Underground one?
20160830 06:16:25< celmin> I dunno the number.
20160830 06:19:15-!- celticminstrel is now known as celmin|sleep
20160830 06:24:21-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160830 06:30:39< wedge009> Oh dear, something committed today seems to have broken the victory conditions dialogue.
20160830 06:32:03< wedge009> Okay, maybe not today, but recently.
20160830 06:35:20< wedge009> celmin: 7a Into the Depths has a static Eye Guardian but it still has 1/1 MP.
20160830 06:36:36-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160830 06:38:25< zookeeper> HttT has the sleeping gryphon... :p
20160830 06:39:10< zookeeper> anyway, why are you asking?
20160830 06:39:29-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20160830 06:39:29-!- wedge010 is now known as wedge009
20160830 06:39:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 06:41:30< wedge009> zookeeper: Oh, good point. Asking because of https://gna.org/bugs/?23720
20160830 06:42:11< wedge009> While people are here - can anyone confirm that current master can't show Victory Conditions properly? Try the first scenarios in HttT and TSG, for example.
20160830 06:43:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160830 06:48:19< wedge009> Gryphon has 9/9 MP, BTW. Must be made to be static via WML.
20160830 06:50:01< zookeeper> sleeping gryphon
20160830 06:51:02< zookeeper> ...too bad my build crashes when trying to start gryphon mountain
20160830 06:54:27< wedge009> Oh yes, you're right.
20160830 06:54:50-!- irker027 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160830 06:54:50< irker027> wesnoth: Charles Dang wesnoth:master 307192c14724 / src/gui/dialogs/lobby/lobby.cpp: MP Lobby: ensure the num games shown label is always updated when applying a fil https://github.com/wesnoth/wesnoth/commit/307192c14724ec9b9a67b76fa66c6d6bd294fe0a
20160830 07:13:34-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 260 seconds]
20160830 07:13:58< shadowm> zookeeper: What multiplayer people?
20160830 07:22:50< zookeeper> shadowm, like those new moderators of ours who seem to hang around
20160830 07:25:21< shadowm> I didn't know you chose mods based on their gameplay expertise.
20160830 07:28:29< zookeeper> no, but they're the closest thing we have right now
20160830 07:30:01< shadowm> Well, if you trust their judgment over yours, that's your call then.
20160830 07:31:05< zookeeper> in this case yeah i'm pretty sure they're better than me at seeing potential problems that may arise from "fixing" the assassin
20160830 07:34:22-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Ping timeout: 255 seconds]
20160830 07:40:39< vultraz> celmin|sleep: you broke formatting in Objective
20160830 07:42:42-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160830 07:42:55< vultraz> wedge009: confirmed the bad markup in objectives
20160830 07:43:02< wedge009> Wow, it's been ages since I've had a black-out.
20160830 07:43:23< wedge009> vultraz: Thanks. I was in the middle of trying to fix it.
20160830 07:45:55< vultraz> any success?
20160830 07:46:55< wedge009> No, only just got power back a few minutes ago. I was looking for how the tagging worked, because I'm still learning how things are done. How do you know it was celmin who broke it?
20160830 07:48:10< vultraz> he just made a change to the objectives.lua file
20160830 07:48:41-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 07:48:42< travis-ci> wesnoth/wesnoth#10609 (master - 307192c : Charles Dang): The build has errored.
20160830 07:48:42< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156138187
20160830 07:48:42-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 07:49:33< vultraz> and reverting his change fixes it
20160830 07:53:25< vultraz> ok, I have a temporary fux
20160830 07:53:27< vultraz> fix
20160830 07:55:38< irker027> wesnoth: Charles Dang wesnoth:master ae6b8f97f256 / data/lua/wml/objectives.lua: Temporary fix for broken Objective dialog formatting (f866039e813b539) https://github.com/wesnoth/wesnoth/commit/ae6b8f97f25689eeb57207fb487f0772711f386c
20160830 07:55:42< vultraz> wedge009: ^
20160830 07:56:26< vultraz> and celmin|sleep ^
20160830 07:57:20< vultraz> the "real" fix is likely somewhere else
20160830 07:57:37< vultraz> this just avoids the issue until it can be applied
20160830 08:09:27-!- atarocch [~atarocch@natmobil.sfa.se] has joined #wesnoth-dev
20160830 08:12:47< wedge009> vultraz: That does resolve it, for now at least.
20160830 08:15:05< irker027> wesnoth: Charles Dang wesnoth:master 3df4539d5888 / src/gui/dialogs/lobby/ (player_info.cpp player_info.hpp): Player Info: cleaned up some unused arguments https://github.com/wesnoth/wesnoth/commit/3df4539d58883c01ba98741491428d8c99739bbd
20160830 08:16:01< wedge009> Should have thought to look in the lua stuff. I was trying to find where the end span and small tags had gone.
20160830 08:28:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 08:32:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160830 08:39:19< irker027> wesnoth: Charles Dang wesnoth:master 3cb8878b9adb / data/ (6 files in 2 dirs): Fix duplicate loading of eras in mp https://github.com/wesnoth/wesnoth/commit/3cb8878b9adbee7fd16d80a157747680cc83565c
20160830 08:45:26< irker027> wesnoth: Charles Dang wesnoth:master 9608e9220999 / data/era_blank.cfg: Removed unnecessary empty era https://github.com/wesnoth/wesnoth/commit/9608e922099989efe3662db4f48063e35bef9497
20160830 08:49:41-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 08:49:42< travis-ci> wesnoth/wesnoth#10610 (master - ae6b8f9 : Charles Dang): The build has errored.
20160830 08:49:42< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156149015
20160830 08:49:42-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 08:53:52-!- Kwandulin [~Miranda@p200300760F4241BF5CB29195B8D86938.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160830 09:07:14-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160830 09:07:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160830 09:07:51-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20160830 09:29:05-!- Kwandulin [~Miranda@p200300760F4241BFA5A7BF0065FBCA0C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160830 09:36:00-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 09:36:01< travis-ci> wesnoth/wesnoth#10611 (master - 3df4539 : Charles Dang): The build has errored.
20160830 09:36:01< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156152595
20160830 09:36:01-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 09:37:53-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has joined #wesnoth-dev
20160830 10:03:27-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 10:03:28< travis-ci> wesnoth/wesnoth#10612 (master - 3cb8878 : Charles Dang): The build has errored.
20160830 10:03:28< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156157398
20160830 10:03:28-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 10:16:24-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 10:20:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160830 10:37:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 10:38:50-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 10:38:51< travis-ci> wesnoth/wesnoth#10613 (master - 9608e92 : Charles Dang): The build has errored.
20160830 10:38:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156158667
20160830 10:38:51-!- travis-ci [~travis-ci@ec2-54-146-3-17.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 10:41:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20160830 10:43:51< irker027> wesnoth: Charles Dang wesnoth:master 14c44d0ea90d / data/gui/window/game_load.cfg src/gui/dialogs/game_load.cpp: Load Game: fixed game list being empty with long filenames on small resolutions https://github.com/wesnoth/wesnoth/commit/14c44d0ea90df3722b4e41df2bb6c9a0cab75940
20160830 10:48:10< irker027> wesnoth: Charles Dang wesnoth:master 40d9f7330657 / src/gui/widgets/listbox.cpp: Attempt to satisfy travis https://github.com/wesnoth/wesnoth/commit/40d9f73306577e23310ae214530db0670c84b46d
20160830 11:12:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 11:17:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160830 11:38:07-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 11:38:08< travis-ci> wesnoth/wesnoth#10614 (master - 14c44d0 : Charles Dang): The build has errored.
20160830 11:38:08< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156183699
20160830 11:38:08-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 12:22:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 12:24:18-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 12:24:19< travis-ci> wesnoth/wesnoth#10615 (master - 40d9f73 : Charles Dang): The build has errored.
20160830 12:24:19< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156184634
20160830 12:24:19-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 12:27:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160830 12:55:05-!- Kwandulin [~Miranda@p200300760F4241BFA5A7BF0065FBCA0C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160830 12:56:17-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160830 12:56:17-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Client Quit]
20160830 12:56:30-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160830 12:57:34-!- aeonchild is now known as enchi
20160830 13:11:07-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160830 13:20:31-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 13:22:39-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160830 13:23:43-!- JyrkiVesterinen [~JyrkiVest@nblzone-242-23.nblnetworks.fi] has quit [Quit: .]
20160830 13:25:00-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160830 13:25:20< mattsc> wedge009: yes, there are units like that all over some UMC campaigns.
20160830 13:25:47< mattsc> That is, they usually have 1 MP and high movecost on all terrain — the reason for that being exactly that bug.
20160830 13:26:16< mattsc> And, in fact, the same is true for the eyestalk in TSG.  It’s placed on cave terrain where it cannot move.
20160830 13:26:35< mattsc> *surrounded by cave terrain
20160830 13:36:44-!- celmin|sleep is now known as celticminstrel
20160830 13:48:26-!- irker027 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160830 13:56:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160830 13:56:37-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev
20160830 13:58:14< mattsc> wedge009, vultraz: btw, the 0 MP units do not attack if there is also no other unit that can get the the enemy adjacent to them.
20160830 13:58:32< mattsc> That’s why sometimes it seems like people cannot reproduce the bug.
20160830 13:59:07< mattsc> there’s an ‘only’ missing in that first sentence
20160830 14:01:41-!- louis94 [~~louis94@91.178.242.113] has quit [Quit: Konversation terminated!]
20160830 14:01:55-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 14:10:39< Elvish_Hunter> celticminstrel: I just told Nobun about the plural forms support in wmlxgettext. Hopefully we'll manage to implement it.
20160830 14:11:23< celmin> http://gna.org/bugs/?11811
20160830 14:11:38< celmin> Elvish_Hunter: ^ The relevant issue report
20160830 14:11:43< celmin> If you missed it.
20160830 14:15:39< Elvish_Hunter> I already got it, thanks.
20160830 14:17:47-!- louis94 [~~louis94@91.178.242.113] has quit [Ping timeout: 265 seconds]
20160830 14:45:09-!- fabi__ [~fabi@176.0.30.136] has joined #wesnoth-dev
20160830 14:48:15-!- fabi_ [~fabi@176.0.30.39] has quit [Ping timeout: 244 seconds]
20160830 14:57:32-!- fabi_ [~fabi@176.5.137.16] has joined #wesnoth-dev
20160830 15:00:46-!- fabi__ [~fabi@176.0.30.136] has quit [Ping timeout: 252 seconds]
20160830 15:02:18-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160830 15:14:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160830 15:15:56< tad_> [ 91%] Building CXX object src/CMakeFiles/wesnoth.dir/gui/dialogs/lobby/lobby.cpp.o virtual memory exhausted: Cannot allocate memory
20160830 15:16:09< celmin> Wow.
20160830 15:16:27< celmin> That's impressive.
20160830 15:16:28< tad_> Might want to consider this. It's solvable .. just need to kill off a few tasks. But it's new.
20160830 15:16:43< celmin> Although I guess you did mention something about a VM earlier...
20160830 15:17:50< tad_> Yep. If I added a swap it would not have died, but it would have been SLLOOOWWWW. I closed my editor and re-ran and it's fine.
20160830 15:18:40< tad_> But it's new in the past couple of days. Since I upgraded gcc last month I've not had memory issues building .. until this.
20160830 15:19:06< tad_> What's odd is it's an 1800 line module and I don't see why it would use twice the memory of any other to compile.
20160830 15:19:30< celmin> I wouldn't expect it to be using many templates either.
20160830 15:20:01< tad_> It's not much of a problem but something to wonder about
20160830 15:21:03< celmin> Is there anything besides templates that might increase memory usage...
20160830 15:21:15 * tad_ shrugs
20160830 15:21:45< tad_> #include <kitchen.sink> would do it. But there's not a huge #include chain at the top.
20160830 15:22:10-!- midzer_ is now known as midzer
20160830 15:22:19< celmin> Well, every include further includes more stuff, though.
20160830 15:22:49-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160830 15:22:52-!- Kwandulin [~Miranda@p200300760F4241BF28724764ECD65B78.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160830 15:23:01< tad_> My first through was a recursive macro or include, or some horrible if statement but nothing lept out when I scanned the source.
20160830 15:23:36< tad_> It's something to wonder .. what changed that suddenly it's taking a LOT more memory to compile.
20160830 15:25:05< celmin> There were some recent changes to data.?pp...
20160830 15:25:18-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 15:25:56< tad_> Anyway .. mainly I'm waiting for make to finish so I can check a change vultraz put in about errors on-screen when opening the editor. I expect it's fine but wanted to check and got an assert failure because I didn't re-make after rebase.
20160830 15:28:25< tad_> I have a scrolling memory-use panel open most of the time. Usually when I'm making BfW it's a saw-tooth. Used to be I could tell when it was linking because would go up up up up BAM and I'd have to clsoe everything to link. New gcc fixed that. Until today on that one source. Everything else is a sawtooth like normal at about 50% use because Chrome is a pig.
20160830 15:30:01-!- atarocch [~atarocch@natmobil.sfa.se] has quit [Ping timeout: 255 seconds]
20160830 15:31:41< vultraz> celmin: yes
20160830 15:31:48< vultraz> mp lobby file
20160830 15:32:01< celmin> I have no idea what you're saying yes to.
20160830 15:32:40< vultraz> [02:25:03]	celmin	There were some recent changes to data.?pp...
20160830 15:34:36< celmin> Now I'm confused why you're saying yes to it.
20160830 15:36:24< tad_> x-post because I want to say it publicly ... You know, I do appreciate y'all working on stuff like this. I still amazes me how CLEAN BfW compiles and with all these niggling little changes, it's running very cleanly, too. Far too many systems get 'clean' simply by not reporting issues. But playing ostrich eventually bites you; good to see a project this large which doesn't.
20160830 15:36:48< celmin> "playing ostrich"?
20160830 15:37:08< tad_> Head in sand. "What I can't see can't hurt me"
20160830 15:37:09 * celmin also wonders what you mean by compiling clean. Free of warnings or something?
20160830 15:37:11< vultraz> head in the sand, maybe?
20160830 15:37:13< vultraz> ah
20160830 15:38:39< tad_> The only messages when compiling are upstream from boost and a long-term known problem they'll eventually fix.  Ignoring that you're compiling at max messages: 0 errors, 0 warnings and generally, now, runing with no warnings or errors, too, for the campaigns I've tested and been working on.
20160830 15:39:08-!- irker234 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160830 15:39:08< irker234> wesnoth: Charles Dang wesnoth:master ce2f274f040a / src/gui/widgets/listbox.cpp: Yet another attempt to satisfy travis https://github.com/wesnoth/wesnoth/commit/ce2f274f040ad851ad5654516ad60f710b869959
20160830 15:39:32< vultraz> I think i understand why travis doesn't like that line now
20160830 15:39:44< vultraz> likely something to do with the fact that cols uses auto
20160830 15:40:15< celmin> That seems unlikely to me...
20160830 15:40:34< celmin> But if you need to change it from auto, the type is config::const_child_itors, I think.
20160830 15:40:52< vultraz> we'll see what travis says
20160830 15:43:17-!- atarocch [~atarocch@88.131.217.34] has joined #wesnoth-dev
20160830 15:43:53 * vultraz is having a hard time deciding what code in the gui1 lobby related to game data needs to be copied to the gui2 version, and which is style differences or does not need to be
20160830 15:45:44 * vultraz is working on porting that 'addons needed' feature
20160830 15:48:27< tad_> Question (maybe I'm out of date) .. why "auto cols" .. do you really need the compiler to deduce the type of cols?
20160830 15:48:57< celmin> Probably laziness and not wanting to type out the full type. >_>
20160830 15:49:34< vultraz> indeed
20160830 15:49:40< tad_> Well, try 'unsigned int cols' and that static_cast and == should be a happy camper?
20160830 15:49:58< celmin> It's not an unsigned int.
20160830 15:50:06< celmin> It's an iterator range.
20160830 15:50:19< tad_> And you wonder why Travis is waxing pedantic?
20160830 15:50:26< celmin> What?
20160830 15:50:38< tad_> No happy about that VALIDATE
20160830 15:50:52< celmin> What are you trying to say.
20160830 15:52:00< tad_> I see and 'auto cols' and 'cols == static_cast()' and a few changes trying to get the static cast right when I wonder if the auto deduction is the issue, not the cast.
20160830 15:52:55< celmin> I haven't actually seen the error on Travis.
20160830 15:53:45< tad_> Nor I but I've seen a couple changes trying to fit the RH side to the 'auto cols' ...
20160830 15:56:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 15:57:36< irker234> wesnoth: ln-zookeeper wesnoth:master c0c696a8d016 / / (17 files in 4 dirs): New Tentacle of the Deep sprite with animated ripples, by Blarumyrran https://github.com/wesnoth/wesnoth/commit/c0c696a8d01623fc7b075adc3896214c115d7b34
20160830 15:57:38< irker234> wesnoth: ln-zookeeper wesnoth:master 90b0d664339a / src/gui/widgets/listbox.cpp: Merge branch 'master' of https://github.com/wesnoth/wesnoth https://github.com/wesnoth/wesnoth/commit/90b0d664339a0fe050eed18cb9c6021c4c41741e
20160830 15:57:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 15:58:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 15:58:33< celmin> Merge commits. :(
20160830 15:58:59< celmin> I can't see the ripples on github. I guess they're almost transparent.
20160830 15:59:19< celmin> Whoa, it looks very different.
20160830 15:59:20< celmin> Nice though.
20160830 15:59:29< vultraz> i should add changelog commits
20160830 15:59:56< celmin> Me too, though I've been doing it periodically already.
20160830 16:02:46< tad_> Might add a checkin script to your local git to remind you ...
20160830 16:02:53< zookeeper> yeah they're very faint.
20160830 16:03:43< zookeeper> (and you can blame git for merge commits)
20160830 16:03:50< vultraz> no
20160830 16:03:58< celmin> Internal changes don't get changelog entries though (unless they're huge, I guess, like SDL2), so that would be more annoying than useful.
20160830 16:04:23< celmin> Again I have no idea what vultraz is talking about.
20160830 16:04:56< vultraz> new heavy infantryman line sprites
20160830 16:05:02< vultraz> more new lisar sprite frames
20160830 16:05:07< vultraz> new gui2 dialogs
20160830 16:05:19< celmin> Sure.
20160830 16:06:28< zookeeper> also someone has (again) not added proper linebreaks in players_changelog
20160830 16:07:50< celmin> You mean for the hard-wrapping?
20160830 16:08:20< zookeeper> i mean the thing that's obvious if you actually look at it
20160830 16:10:27-!- gfgtdf [~chatzilla@x4e32b118.dyn.telefonica.de] has joined #wesnoth-dev
20160830 16:12:51-!- JyrkiVesterinen [~JyrkiVest@89-166-103-195.bb.dnainternet.fi] has joined #wesnoth-dev
20160830 16:13:31< vultraz> i see nothing
20160830 16:15:14-!- louis94 [~~louis94@91.178.242.113] has quit [Ping timeout: 265 seconds]
20160830 16:16:15< tad_> zookeeper: I have a PR up to improve the AI ignoring villages on AToTB. Quick and clean and if I'd have known how to do it before the big merge last week, it would have been in there ...
20160830 16:17:27< tad_> celmin: I have a PR up we talked about for [modify_unit] .. I've moved away from the [filter_recall] work but it seemed to help .. want me to cancel it until I get back to that area?
20160830 16:17:51 * tad_ is trying to clean up is local repo ...
20160830 16:19:54< celmin> Why would you cancel it?
20160830 16:20:43< tad_> Might be a while before I get back to it. I can store it off and we can re-visit the idea later.
20160830 16:21:39< tad_> I've been deleting local branches because it was getting hard to remember what was important and what was junk as I tried to figure out what I wanted to do ...
20160830 16:22:38< celmin> I still don't see how that leads to cancelling it though.
20160830 16:23:09< tad_> Taking stuff off the table, is all.
20160830 16:23:42< tad_> I suppose it's fine. Just not sure we need it right now.
20160830 16:23:46< celmin> I don't know any reason not to just merge it.
20160830 16:24:12< celmin> I didn't merge it already mainly to give other people a chance to look at it, but it seems like that's not happening...
20160830 16:24:23< tad_> Does it ever?
20160830 16:24:27< gfgtdf> vultraz: the travis wanrings came becasue up to boost 1.56 iterator_range::size() retnred an signes value (since it returns end() -begin() whihc nca be neagtive)
20160830 16:24:30< tad_> People look when we break it ...
20160830 16:24:49< vultraz> ahh
20160830 16:24:56< zookeeper> tad_, #755?
20160830 16:25:10< celmin> gfgtdf: If that's the case, why did they change it to unsigned?
20160830 16:25:25< tad_> zookeeper: yes 755
20160830 16:25:39< celmin> vultraz, gfgtdf: Any objections to merging PR759?
20160830 16:25:44< zookeeper> tad_, right, i wasn't sure if you were gonna do something else to that. i'll merge...
20160830 16:25:46< gfgtdf> celmin: mostlikley to make it eaisert to use in as a drop in for other collections, but not sure.
20160830 16:26:07< vultraz> celmin: no
20160830 16:26:08< gfgtdf> there are soem branches at github like Vultraz-event_handling_fixes, boost_trimming and tc_deprecate please delete remote branches if tehy have been merged or abdonned.
20160830 16:26:10< vultraz> objection
20160830 16:26:18< celmin> Well, logically it makes more sense for it to return abs(end() - begin()) I guess...
20160830 16:27:04< tad_> When I pulled down the master repo I deleted ALL branches except master. Made it a LOT cleaner .. there's a lot of abandoned stuff there, looks like.
20160830 16:27:33< celmin> tad_: I did that too.
20160830 16:27:33< irker234> wesnoth: Gregory A Lundberg wesnoth:master 6279145c159c / data/campaigns/Two_Brothers/scenarios/02_The_Chase.cfg: AToTB S02 Improve AI https://github.com/wesnoth/wesnoth/commit/6279145c159c2177cc1d27c1b85d4da44e485208
20160830 16:27:35< irker234> wesnoth: Lari Nieminen wesnoth:master 8121eef2a044 / data/campaigns/Two_Brothers/scenarios/02_The_Chase.cfg: Merge pull request #755 from GregoryLundberg/GL_AToTB_S02_improve_ai https://github.com/wesnoth/wesnoth/commit/8121eef2a0442063b94622f4b2e1f99b287a05dc
20160830 16:27:35< gfgtdf> celmin: hmm that pr look liek it could clash with a possible plan to make [modify_unit] use diewctly lua unit manipulation instead of store+unstore,
20160830 16:27:40< celmin> Both on my fork and in my local repo
20160830 16:27:55< gfgtdf> celmin: but i dont really casre that much
20160830 16:28:10< celmin> gfgtdf: It would clash with that if that plan had already been started, sure.
20160830 16:31:24< irker234> wesnoth: Gregory A Lundberg wesnoth:master 82c3b0114d43 / changelog data/lua/wml/modify_unit.lua: Add mode=replace to [modify_unit] https://github.com/wesnoth/wesnoth/commit/82c3b0114d4381ea53d2fb2ca93b3775418ef7d4
20160830 16:31:26< irker234> wesnoth: Celtic Minstrel wesnoth:master b24fdbcbe196 / changelog data/lua/wml/modify_unit.lua: Merge pull request #759 from GregoryLundberg/GL_modify_unit_mode https://github.com/wesnoth/wesnoth/commit/b24fdbcbe196c20da3c0c0f4daa30d2037204b5a
20160830 16:32:36< tad_> If the talk about not using store/unstore in [modify_unit] was about that PR .. YES it DEPENDS upon using unstore to create a new instance of C++ unit because that's the only way to replace/update [filter_recall]
20160830 16:33:18< tad_> To do away with unstore, you'll need to update unit so the field can be changed after instantiation
20160830 16:33:27-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 16:33:28< travis-ci> wesnoth/wesnoth#10616 (master - ce2f274 : Charles Dang): The build has errored.
20160830 16:33:28< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156261455
20160830 16:33:28-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 16:33:34< celmin> Didn't I do that? Maybe I made it read-only.
20160830 16:33:42< celmin> Oh, right/
20160830 16:33:55< celmin> What I did is add it to the Lua.
20160830 16:33:59< tad_> Or maybe I didn't read it right ... 
20160830 16:34:08< celmin> But I think only in the __index metatable.
20160830 16:34:37< tad_> note that in the C++ the name changes from filter_recall to recall_filter at one point, too.
20160830 16:35:01< celmin> Anyway, if updating the C++ is necessary to support that change to [modify_unit], then the C++ can be updated.
20160830 16:35:24< celmin> It's not like we're trying to work around bugs in the engine using just Lua.
20160830 16:35:37< celmin> ^bugs/insufficiencies
20160830 16:35:57< tad_> Yes, just raising my hand that we'll want to consider all the fields one might change with [modify_unit] and not using store/unstored
20160830 16:36:15< celmin> Of course.
20160830 16:36:38< celmin> I'm still kinda unsure if I actually want this change though.
20160830 16:36:51< tad_> Which?
20160830 16:37:07< celmin> It's good to avoid store/unstore, but if it means you need to update [modify_unit] every time some new unit attribute is added, that's not good.
20160830 16:38:16< tad_> What bothers me with [modify_unit] is those two special cases in the if-logic .. otherwise it looked pretty insensitive to what was being changed.
20160830 16:38:27< celmin> Special cases?
20160830 16:38:34< celmin> The modifications and the effects?
20160830 16:38:39< tad_> yea
20160830 16:38:54< tad_> I punted and didn't let mode=replace touch them
20160830 16:39:16< celmin> Well, they aren't things that appear in a unit.
20160830 16:39:37< celmin> Modifications would normally go under [modifications], and effects within that.
20160830 16:40:14< celmin> ^within the individual modifications
20160830 16:40:40< tad_> How are they not in unit? Are they not sub-tags?
20160830 16:40:48< celmin> They are not subtags of unit, no.
20160830 16:40:54< celmin> That's why they're special cases.
20160830 16:41:21-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160830 16:41:26< celmin> [object], [trait], and [advancement] can appear under [unit][modifications], so in principle [modify_unit] could change them without the special case.
20160830 16:41:27< tad_> OK, so I'm confused .. why is [modify_unit] changing them, then? Seems like crossing resposibility boundaries to me
20160830 16:41:48< celmin> However, I think the reason for the special case is to support an "append" mode for modifications.
20160830 16:42:18< celmin> To do that using [modify_unit][modifications], you'd need to know exactly what sorts of modifications the unit has.
20160830 16:43:23< celmin> [modify_unit][effect] uses the effect to describe a change to the unit, the same way modifications use effects.
20160830 16:43:32< tad_> Hmm. Well, that was precisely the issue with [filter_recall] .. I needed to know what to remove instead of appending .. so mode=replace clears it first.
20160830 16:43:41< celmin> The difference being that the change is not recorded to be reapplied at advancement time.
20160830 16:43:57< celmin> The default is merge, not append.
20160830 16:44:35< tad_> Well, anyway .. those are areas I've not learned so I just punted and mode=replace does not work for them.
20160830 16:45:04< celmin> Yeah.
20160830 16:47:08< tad_> What struck me as odd was you and I had been working on this area and a few days later a forum topic came up which had a lot of store/unstore work which filter_recall and mode=replace were the best solution if you're on 1.13.5+dev .. was sorta strange that others had the same issue with 1.10 and 1.12 at the same time, is all.
20160830 16:48:29< tad_> [ 77%] Building CXX object src/CMakeFiles/wesnoth-core.dir/tstring.cpp.o /home/lundberg/wesnoth/src/tstring.cpp: In constructor ‘t_string_base::t_string_base(const string&, const string&, int, const string&)’: /home/lundberg/wesnoth/src/tstring.cpp:295:39: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]   *reinterpret_cast<int32_t*>(count_val) = count;
20160830 16:48:40< celmin> I could probably make use of [filter_recall] in my campaign, too.
20160830 16:49:12< celmin> Though, not sure if there's a way to write a filter that matches "anything who started as unit type X" when there are units that started as advancements of X.
20160830 16:49:37< tad_> My proof-of-concept for TSG says it will solve a LOT of issues for when units need to come and go or you're changing leaders a lot .. DM and LoW come to mind
20160830 16:49:43< celmin> I suppose you could do a formula filter on the unit ID, but, well, that's probably not future-proof.
20160830 16:50:00< Ravana_> started, or could have started with crossing unit line?
20160830 16:50:09< celmin> Ravana_: What?
20160830 16:50:25< Ravana_> A advanced to C,D
20160830 16:50:33< Ravana_> B advanced to D,E
20160830 16:50:39< Ravana_> how to treat D
20160830 16:50:45< celmin> Ah, no, it's not that.
20160830 16:50:47< tad_> [unit][variables] can help. I used them in my proof-of-concept to mark "Sir Gerrick's Army"
20160830 16:51:02< celmin> It's just X advances to A,B.
20160830 16:51:19< celmin> The issue is that there are probably a few units who started as A or B.
20160830 16:51:28< celmin> tad_: Yeah, that's probably the best way.
20160830 16:54:32< tad_> In fact, except for cases where we need to grab something from a unit (which is another issue ..) I would posit that if you see a lot of store/unstore work, especially when it crosses scenario boundaries [filter_recall] should be the preferred solution. Things are so much better if you don't have all those variables laying around .. like units get healed properly, can move when recalled by WML .. etc.
20160830 16:57:34-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Remote host closed the connection]
20160830 16:57:56< tad_> Plus one of the major issues getting debug/choose_level to work if groking which units need to be on the recall list and which in variables .. which goes away if you use [filter_recall] and keep them all on the recall list all the time.
20160830 16:58:34-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160830 17:04:55< vultraz> celmin: btw, did you see my commit patching objectives?
20160830 17:05:10< vultraz> is there some issue in your tstring commit?
20160830 17:05:32< celmin> It worked when I tested it.
20160830 17:05:45< celmin> That was in the Lua console, mind you.
20160830 17:06:10< celmin> So I never saw that broken formatting.
20160830 17:07:25< tad_> celmin: your commit 20 hours ago triggered a -Wstrict-aliasing warning
20160830 17:07:37< celmin> Which commit?
20160830 17:07:49< vultraz> i assume the tstring one
20160830 17:07:55< celmin> And where?
20160830 17:07:56< tad_> Add support for gettext plurals to Lua API
20160830 17:08:03< tad_> I think that's the culprit
20160830 17:08:22< celmin> What's strict aliasing anyway?
20160830 17:09:04< tad_> I think you cast away const-ness .. but I'd have to look closer to be sure
20160830 17:09:15< celmin> What line?
20160830 17:09:30< tad_> [11:48] <tad_> [ 77%] Building CXX object src/CMakeFiles/wesnoth-core.dir/tstring.cpp.o /home/lundberg/wesnoth/src/tstring.cpp: In constructor ‘t_string_base::t_string_base(const string&, const string&, int, const string&)’: /home/lundberg/wesnoth/src/tstring.cpp:295:39: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]   *reinterpret_cast<int32_t*>(count_val) = count;
20160830 17:10:24< JyrkiVesterinen> celmin, tad_: strict aliasing means that a memory area can only be pointed to by pointers/references of the same type.
20160830 17:10:49< JyrkiVesterinen> Compilers assume that pointers/references of different types don't alias, i.e. point to the same memory area.
20160830 17:10:49< celmin> I suppose I could do it the way the textdomain ID is handled and manually break the integer into chunks.
20160830 17:10:52< tad_> Well, it's been 15 or so years since I did any serious C++
20160830 17:11:04< JyrkiVesterinen> It allows some significant performance optimizations.
20160830 17:11:30< JyrkiVesterinen> I think triggering a strict aliasing warning on a reinterpret_cast is quite silly, though.
20160830 17:11:44< celmin> Or I could use a union?
20160830 17:11:50< celmin> I dunno.
20160830 17:12:00< JyrkiVesterinen> Yes, a union would work.
20160830 17:12:10< celmin> Can that be the cause of the broken formatting though?
20160830 17:12:27< tad_> I usually presume the gcc folk have a good reason for all that stuff. Sometimes they're wrong but usually they have a good point
20160830 17:12:37< celmin> vultraz: What was the broken formatting, anyway?
20160830 17:12:43< celmin> Since I never saw it.
20160830 17:12:50< JyrkiVesterinen> Here is a commit where I replaced reinterpret_cast with a union: https://github.com/wesnoth/wesnoth/commit/4c2338d61618f817735d60e34ec98232ce4fa45f
20160830 17:12:57< vultraz> celmin: it seemed it didn't append the closing tags
20160830 17:13:27< celmin> So somehow the closing tag was skipped?
20160830 17:13:33< celmin> That suggests the problem is in the walker.
20160830 17:13:56< JyrkiVesterinen> Here's a good read about strict aliasing rules: http://blog.regehr.org/archives/1307
20160830 17:14:02< celmin> Because from the Lua code, the ending tags are clearly appended.
20160830 17:14:13< vultraz> yes
20160830 17:14:17< vultraz> but it didn't
20160830 17:14:28< vultraz> i had to move them directly into the format call
20160830 17:14:32< celmin> It probably did append them.
20160830 17:14:38< celmin> But had a null character or something.
20160830 17:14:43< vultraz> perhaps
20160830 17:14:53< vultraz> i assume you will fix this :)
20160830 17:15:13< celmin> I'll look into it after I finish with matrix placement. Though I haven't actually touched that yet today.
20160830 17:15:27< tad_> Oh .. vultraz I been meaning to mention .. I get errors from [inspect] on the console. Soemthing about widths not being right. It's not real consistent and seems to have to do with an edge condition.
20160830 17:15:41< celmin> At least I finally decided last night on how to calculate the best size.
20160830 17:15:53< celmin> The layout already looks much better.
20160830 17:16:02< vultraz> tad_: gui errors?
20160830 17:16:02< celmin> But it's not placing them properly yet.
20160830 17:16:03< vultraz> or wml errors
20160830 17:17:17< tad_> vultraz: gui errors, I think. I can't get it to reproduce at will because I usually dont notice it until its too late .. but my impression is it has a problem when there's no units for a list .. or only one .. or maybe it's variables .. it does seem to be an edge condition like that tho. 
20160830 17:17:31< vultraz> hm hm hm hm hm
20160830 17:17:35< vultraz> hhhhmmmmmmm
20160830 17:17:40 * vultraz ponders
20160830 17:17:42< tad_> Some long cryptic this-in-this-in-this message which is very hard to read
20160830 17:17:55< vultraz> anything about 'cannot place grid'?
20160830 17:18:11< tad_> Hmm. Let me see if I can trigger one.  Hang on.
20160830 17:20:10 * tad_ sighs
20160830 17:20:44< tad_> Of course not. Well, I'll grab one when next I see it. IIRC it has to do with a width being a couple chars wider than another.
20160830 17:21:28-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160830 17:23:30-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 17:23:31< travis-ci> wesnoth/wesnoth#10617 (master - 90b0d66 : ln-zookeeper): The build has errored.
20160830 17:23:31< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156266356
20160830 17:23:31-!- travis-ci [~travis-ci@ec2-54-81-19-76.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 17:27:34-!- jswensen [~jswensen@s48075040.temp.wsu.edu] has joined #wesnoth-dev
20160830 17:28:31< celmin> Hi, did you figure out the fontconfig issue?
20160830 17:30:11< jswensen> Yes, I think so. I just needed to set the FONTCONFIG_PATH and FONTCONFIG_FILE environment variables to point to the .app bundle root.
20160830 17:30:26< jswensen> Still trying to get it to find all the config files and set up the data directories.
20160830 17:31:43< celmin> I have no idea if this'll help, but the OSX version places them in .app/Resources/. Also, there's some compile-time way to set the default search path for resources (which can be overridden from the command-line).
20160830 17:31:53< celmin> ^ for data
20160830 17:32:30< jswensen> iOS is a little different. Because they enforce sandboxing, it is explicitly forbidden to write to the .app directory.
20160830 17:32:51< celmin> It doesn't write to data/
20160830 17:32:56< celmin> Only userdata/
20160830 17:33:08< celmin> (Which isn't usually called userdata/)
20160830 17:33:36< celmin> OSX places the userdata in Application Support.
20160830 17:40:24< jswensen> I guess I am a little confused where the data dir is specified. I see a bunch going on in filesystem_boost.cpp that deals with user_data_dir, user_config_dir, and cache_dir, which I’m assuming should all point to the iOS equivalent of Application Support, but where do I point the app to the data directory (which I am assuming will ensure that fonts, data, images, and sounds are all located correctly?
20160830 17:40:52-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160830 17:43:20< jswensen> I see that in the startup code for OSX that they actually change directory to the .app’s resource directory (chdir([[[NSBundle mainBundle] resourcePath] fileSystemRepresentation]);). I will have to check and see whether that is actually allowed on iOS. 
20160830 17:43:37< JyrkiVesterinen> The data directory is game_config::path.
20160830 17:43:57< JyrkiVesterinen> The code that tries to find it automatically, if not specified in the command line, starts here: https://github.com/wesnoth/wesnoth/blob/master/src/wesnoth.cpp#L1044
20160830 17:57:23< jswensen> It looks like that is all sorted out now. Next problem is a null pointer dereference deep within the guts of pango_fc. The original pango call is pango_layout_get_pixel_extents from font::ttext::recalculate. 
20160830 17:57:35< celmin> Oh my.
20160830 17:57:48< jswensen> From the OSX version, it looks like I might need to set some environment variables for pango also.
20160830 17:58:45< jswensen> They have “setenv ("PANGO_RC_FILE", "./pangorc", 1);
20160830 17:58:46< jswensen> 	setenv ("PANGO_SYSCONFDIR", ".", 1);
20160830 17:58:47< jswensen> 	setenv ("PANGO_LIBDIR", ".", 1);”, though I don’t actually see a pangorc file anywhere in the .app or Application Support
20160830 17:59:17< celmin> I seem to call it being removed.
20160830 18:00:01< celmin> So I wonder if maybe those lines are now obsolete...
20160830 18:00:28< celmin> Is that in main()?
20160830 18:00:55< jswensen> Well, they had it in applicationDidFinishLaunching, so close to being main().
20160830 18:01:07< celmin> Oh right
20160830 18:01:08< jswensen> It actually preceded the call to wesnoth_main
20160830 18:02:27< celmin> (Until recently there was a pangorc in projectfiles/XCode/Resources)
20160830 18:05:46< jswensen> Also, I noticed that the fonts.conf for OSX is different in the projectfiles/XCode/Resources that the one in the data directory. I currently am copying the default one from the data directory to the app bundle. I should probably use the OSX one.
20160830 18:05:48< jswensen> ?
20160830 18:05:49-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 18:05:50< travis-ci> wesnoth/wesnoth#10618 (master - 8121eef : Lari Nieminen): The build has errored.
20160830 18:05:50< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156275052
20160830 18:05:50-!- travis-ci [~travis-ci@ec2-54-92-142-249.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 18:09:16< jswensen> Or take a look at how they did it in the old wesnoth_ios project. It didn’t seem to spit out an error about the fonts.conf file that I did include, but it is possible that it was structurally OK, but isn’t pointing to the right location.
20160830 18:09:58< celmin> I'm not sure which one you should use.
20160830 18:17:41< jswensen> Well, the one that is in the fonts directory doesn’t even point to a font directory, so I suspect that I probably need to use one like the OSX fonts.conf and change folder references to point to the bundled fonts directory.
20160830 18:22:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 18:50:08-!- travis-ci [~travis-ci@ec2-54-80-189-138.compute-1.amazonaws.com] has joined #wesnoth-dev
20160830 18:50:09< travis-ci> wesnoth/wesnoth#10619 (master - b24fdbc : Celtic Minstrel): The build has errored.
20160830 18:50:09< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/156276058
20160830 18:50:09-!- travis-ci [~travis-ci@ec2-54-80-189-138.compute-1.amazonaws.com] has left #wesnoth-dev []
20160830 18:51:28-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Ping timeout: 264 seconds]
20160830 18:51:28-!- loonycyborg [~loonycybo@wesnoth/developer/loonycyborg] has quit [Ping timeout: 264 seconds]
20160830 18:52:04-!- APic [apic@apic.name] has quit [Ping timeout: 264 seconds]
20160830 18:53:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 18:55:31-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: No route to host]
20160830 18:55:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 18:56:04-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160830 18:57:04-!- TheJJ [~rofl@ipbcc36896.dynamic.kabel-deutschland.de] has quit [Ping timeout: 260 seconds]
20160830 18:57:36-!- loonycyborg [~loonycybo@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20160830 19:01:01 * tad_ spent most of the morning and afternoon cleaning up his local repo, rebasing all those HttT PRs and doing general cleanup. He's really avoiding actual work.
20160830 19:04:00-!- APic [apic@apic.name] has joined #wesnoth-dev
20160830 19:05:10-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20160830 19:07:27< celmin> Sounds familiar. >_>
20160830 19:25:34-!- jswensen [~jswensen@s48075040.temp.wsu.edu] has quit [Quit: jswensen]
20160830 19:31:32-!- irker234 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160830 19:37:44-!- gfgtdf [~chatzilla@x4e32b118.dyn.telefonica.de] has quit [Ping timeout: 244 seconds]
20160830 19:40:08-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160830 19:43:26-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20160830 19:43:27-!- wedge010 is now known as wedge009
20160830 19:51:18-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160830 19:52:31< vultraz> ugh this is tedious 
20160830 19:52:51< celmin> ?
20160830 19:53:41< vultraz> pushing around code for the 'download addons' stuff in the lobby
20160830 19:56:11< zookeeper> hrhm. minor annoyance: if an unselected unit levels via an [unstore_unit] (and maybe in other cases too, i dunno), it won't be selected and thus i don't easily see its current stats, only its advancements' stats. i'd have to go the unit help and through the "advances from" item.
20160830 19:56:43< zookeeper> (if it was selected, i'd see it in the right panel alongside the advancement dialog)
20160830 20:01:18-!- JyrkiVesterinen [~JyrkiVest@89-166-103-195.bb.dnainternet.fi] has quit [Quit: .]
20160830 20:05:38-!- mjs-de [~mjs-de@x4e31abbc.dyn.telefonica.de] has joined #wesnoth-dev
20160830 20:10:56-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160830 20:12:07< zookeeper> the status table (both new and old) doesn't let me see side colors when the sides are leaderless. i can't tell if for example green and purple are allied or not, unless i abuse my inside knowledge of the typical color order (green 3, purple 4).
20160830 20:12:48< vultraz> why would they be allied by color..
20160830 20:14:00< zookeeper> no, i mean i can't look at the team names and see whether they're allied because i don't know which side in the status table is which color.
20160830 20:14:33-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160830 20:14:37< zookeeper> i mean sure i guess it's a fringe issue
20160830 20:17:07< zookeeper> you'd need to have several sides on several teams and then some units for which it's not obvious which team they belong to... so, yeah, not really a problem after all.
20160830 20:17:49< vultraz> do you want me to colorize the team names?
20160830 20:18:17< zookeeper> probably not, since it'd help people confuse colors and teams
20160830 20:22:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 20:30:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 20:35:14-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160830 20:37:02-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 20:48:10-!- louis94 [~~louis94@91.178.242.113] has quit [Quit: Konversation terminated!]
20160830 20:48:29-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 20:49:36-!- louis94 [~~louis94@91.178.242.113] has quit [Read error: Connection reset by peer]
20160830 20:50:39-!- louis94 [~~louis94@91.178.242.113] has joined #wesnoth-dev
20160830 20:57:08-!- mjs-de [~mjs-de@x4e31abbc.dyn.telefonica.de] has quit [Remote host closed the connection]
20160830 21:06:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 21:12:04< vultraz> blah
20160830 21:12:09< vultraz> in must catch bad_enum_cast
20160830 21:12:39< vultraz> i*
20160830 21:13:16< vultraz> oh wait
20160830 21:13:26< vultraz> there's a no-exception version
20160830 21:13:28< vultraz> praise be
20160830 21:21:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 21:22:06< vultraz> i can't tell if every scenario i test has no dependencies..
20160830 21:22:09< vultraz> or my code is broken
20160830 21:25:06< vultraz> ah
20160830 21:25:07< vultraz> it works!
20160830 21:26:10-!- Kwandulin [~Miranda@p200300760F4241BF28724764ECD65B78.dip0.t-ipconnect.de] has quit [Quit: Kwandulin]
20160830 21:35:37-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 21:39:31-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160830 21:42:51-!- Appleman1234_ [~Appleman1@119.104.100.118] has joined #wesnoth-dev
20160830 21:42:51-!- Appleman1234 [~Appleman1@KD119104105128.au-net.ne.jp] has quit [Disconnected by services]
20160830 21:42:52-!- Appleman1234_ is now known as Appleman1234
20160830 21:44:19-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160830 21:51:45-!- louis94 [~~louis94@91.178.242.113] has quit [Quit: Konversation terminated!]
20160830 22:04:07-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 22:06:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 22:10:02-!- jswensen [~jswensen@s48075040.temp.wsu.edu] has joined #wesnoth-dev
20160830 22:10:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160830 22:18:47-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160830 22:23:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 22:32:30-!- atarocch [~atarocch@88.131.217.34] has quit [Quit: Leaving]
20160830 22:39:29-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Read error: Connection reset by peer]
20160830 22:39:35-!- celmin [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160830 22:39:56-!- celticmistral [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160830 22:41:51-!- celmin is now known as celticminstrel
20160830 22:49:22-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160830 22:50:33< tad_> When you re-enable fog at the end of a turn the End Turn button remains grayed/disabled until you click someplace on the map.
20160830 22:51:11< vultraz> BLAH
20160830 22:52:36< tad_> I noted it testing TSG "Long March" which has all that WML path exposing stuff. It may be related to that, but I think not. I'll test someplace else in a moment.
20160830 22:52:49< vultraz> nah, it wouldn't be
20160830 22:53:24< tad_> Still, best to be sure. It's a quick test.
20160830 22:54:17< tad_> OK. It's generic. Also happens in TSG S05.
20160830 22:54:50< tad_> Shroud has same issue.
20160830 22:56:47 * vultraz ponders how to rewrite this string
20160830 22:56:48< vultraz> "You might suggest to them that they update their add-ons." 
20160830 22:57:06< tad_> s/to them that//
20160830 22:57:17< vultraz> good enough
20160830 22:57:29< vultraz> thanks
20160830 23:05:44< tad_> zookeeper: I have updated the TSG PR with the changes to everything except the conversation/objective changes for S05.
20160830 23:08:07-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 255 seconds]
20160830 23:11:30< tad_> Also, I have another bug but it's easier to ref my issue: https://github.com/GregoryLundberg/wesnoth/issues/32 .. I passed it to gfgtdf as a commit message, he may be working it, but the comment was that commit didn't cause it so I dunno
20160830 23:19:07-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160830 23:19:13-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20160830 23:19:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160830 23:21:21-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160830 23:26:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160830 23:34:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 23:35:33-!- irker488 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160830 23:35:33< irker488> wesnoth: Charles Dang wesnoth:master 432fac588e69 / src/ (7 files in 3 dirs): MP Lobby: port 'missing addon requirements' interface from GUI1 lobby https://github.com/wesnoth/wesnoth/commit/432fac588e69f99d34e3020fd6e0ca9ac05a9beb
20160830 23:35:34< irker488> wesnoth: Charles Dang wesnoth:master 8905b9e84c9f / src/gui/dialogs/multiplayer/mp_create_game.cpp: MP Create: provide defaults to prevent crashes if an empty pref field was passed https://github.com/wesnoth/wesnoth/commit/8905b9e84c9ffd3ed7c5bb59a22e0f4d3163ca1e
20160830 23:35:35< irker488> wesnoth: Charles Dang wesnoth:master 7ec92eb754fe / src/game_initialization/create_engine.cpp: Create Engine: if only one difficulty is defined, use it without a dialog prompt https://github.com/wesnoth/wesnoth/commit/7ec92eb754fee2f280c5e081dfb9e892f3670195
20160830 23:39:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160830 23:39:17< vultraz> and yes, i made sure to add the new argument to the tests
20160830 23:39:25< vultraz> argument to lobby_info
20160830 23:42:14< celticmistral> Using std::max on an enum… :|
20160830 23:43:07< celticmistral> Why are you asserting false in the CANNOT_SATISFY case?
20160830 23:45:50< celticmistral> There are very few cases where assert(false) is reasonable, and that's definitely not one of them.
20160830 23:49:58< vultraz> that's copied code
20160830 23:50:12< celticmistral> Ah.
20160830 23:50:12< vultraz> i only did the minimum to get it working right in the new lobby
20160830 23:50:59< vultraz> is there a problem with using std::max on an enum?
20160830 23:51:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160830 23:51:19< celticmistral> Well...
20160830 23:51:30< celticmistral> It's just...
20160830 23:51:47< celticmistral> Weird, I guess?
20160830 23:54:29< vultraz> hmm
20160830 23:54:35< vultraz> there's a have_all_mods flag...
20160830 23:54:42< celticmistral> ?
20160830 23:54:49< vultraz> this should probably be interfaced with the popup..
20160830 23:55:34< vultraz> 				const config &mod = game_config.find_child("modification", "id", cfg["id"]);
20160830 23:55:35< vultraz> 				if(!mod) {
20160830 23:55:37< vultraz> 					have_all_mods = false;
20160830 23:55:38< vultraz> 					break;
20160830 23:55:40< vultraz> 				}
20160830 23:55:41< vultraz> celticminstrel: that
20160830 23:55:54< vultraz> i should probably set addons_outcome to something there..
20160830 23:56:00< vultraz> (and use mod.empty())
20160830 23:56:47-!- TheJJ [~rofl@ipbcc36896.dynamic.kabel-deutschland.de] has joined #wesnoth-dev
20160830 23:58:54< vultraz> but should it be CANNOT_SATISFY..
20160830 23:58:59< vultraz> or NEEDS_DOWNLOAD?
20160830 23:59:47< celticmistral> I thought find_child returned a pointer.
--- Log closed Wed Aug 31 00:00:13 2016