--- Log opened Sat Nov 17 00:00:37 2018 20181117 00:27:36<+wesdiscordbot> @mattsc it's now default_ai_algorithm instead of the existing ai_algorithm? 20181117 00:28:33-!- sevu [~sevu@p5485415E.dip0.t-ipconnect.de] has quit [Remote host closed the connection] 20181117 00:36:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20181117 00:39:33<+wesdiscordbot> @sevu Yes, it is for the default ai_algoritm. Just as it is [default_config], that's a deliberate choice. 20181117 00:40:29<+wesdiscordbot> Note that it is also in parallel to [default_config] rather than inside it, which makes it much easier to deal with given how the code combines [ai] tags. 20181117 00:41:23<+wesdiscordbot> For all "normal" [ai] tags it remains just ai_algorithm. 20181117 00:44:06<+wesdiscordbot> Anyway, there's no clear advantage to doing it one way or the other (except that have just ai_algorithm= inside [default_config] would make it more complex to deal with given how the C++ code works), so I chose to make it clear through the name of the parameter what it is. 20181117 00:44:58-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20181117 01:11:33< celmin|away> IME not keeping WML is pretty much a death flag for 2.x. Maybe not if you have an auto-converter, but even then... 20181117 01:11:33< celmin|away> gfgtdf: If you're around later, we can talk then? Or tomorrow? 20181117 01:11:34< celmin|away> @Pentarctagon If 2.x is using YAML, then it has macro-equivalents. 20181117 01:12:49<+wesdiscordbot> You were asked to stop it with the hostility towards series 2. 20181117 01:13:50<+wesdiscordbot> Throwing around crap like "death flag" is needless negativity. 20181117 01:14:54<+wesdiscordbot> celmin|away: my understanding is that those are not usable across multiple files 20181117 01:15:04<+wesdiscordbot> in which case, it really doesn't 20181117 01:15:04<+wesdiscordbot> You're not even willing to help make the project work, so stop trying to make it seems like an inevitable failure. 20181117 01:15:41<+wesdiscordbot> not to mention more wesnoth-specific things like #ifver 20181117 01:27:34< celmin|away> @Pentarctagon Ah yes, you have a point there. They probably aren't usable across multiple files. 20181117 01:30:07-!- celmin|away is now known as celticminstrel 20181117 02:00:32<+wesdiscordbot> @mattsc the linked commit … I'm a bit confused, it's shown as part of the wesnoth repo, but not to be on the master branch and I think not any else two…I couldn't cherry-pick it, hat do download it as patch 20181117 02:02:19<+wesdiscordbot> I haven't made much testing on it (for 1.14), I think it needs more to see of there's a situation where it causes problems 20181117 02:03:20<+wesdiscordbot> what might be interresting for you 20181117 02:03:21<+wesdiscordbot> https://cdn.discordapp.com/attachments/259976436490829825/513172257271709696/Alternate_Core.zip 20181117 02:04:00-!- irker196 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181117 02:08:55< celticminstrel> Is anyone at all opposed to merging PR 2664? 20181117 02:08:56< celticminstrel> https://github.com/wesnoth/wesnoth/pull/2664 20181117 02:09:53<+wesdiscordbot> @sevu It's still part of a PR at the moment: https://github.com/wesnoth/wesnoth/pull/3723 20181117 02:11:13< buhman> I'm a bit of a lua/c++ noob; I see the gui2::window object lifetime is managed by std::unique_ptr, I think this is half of the reason behind managing "scoped_dialog::current". How does memory management work if a gui2::window were passed to lua as userdata instead? 20181117 02:13:08<+wesdiscordbot> @sevu Sorry, what do you want me to do with those alternate core files? 20181117 02:13:38< buhman> also is there any code that works like this? 20181117 02:15:11<+wesdiscordbot> Was an idea how to get the experimental AI more popular to roll it out as default someday 20181117 02:15:49<+wesdiscordbot> Ah, okay. Obviously I like that idea. 👍 20181117 02:16:33<+wesdiscordbot> Are there any plans (or just ideas of plans) as to when 1.14.6 might come out? 20181117 02:17:02< celticminstrel> buhman: In that case, you probably need to include a __gc method in the dialog's metatable that will free the memory. 20181117 02:17:14< celticminstrel> Unless you can get Lua to allocate enough memory for the entire dialog. 20181117 02:17:26< celticminstrel> But even in that case you'll need a __gc method to call the destructor. 20181117 02:17:32<+wesdiscordbot> Nothing planned so far 20181117 02:17:33<+wesdiscordbot> I'm just asking whether I should try to get that PR out sooner rather than later. So far I have not felt the need to hurry. 20181117 02:17:58< celticminstrel> FTR, by "enough memory for the entire dialog" I basically mean new(L) gui2::window 20181117 02:18:20< celticminstrel> (And make sure you include the header file that defines the custom new operator, otherwise that will overwrite the Lua kernel and break everything.) 20181117 02:18:27<+wesdiscordbot> Either way, I probably won't try to include all the C++ changes in the references issues, but do "some of then" and then merge. 20181117 02:18:30< celticminstrel> (IIRC it's lua_common.h.) 20181117 02:18:36<+wesdiscordbot> Not sure yet what "some of them" means. 20181117 02:19:02< celticminstrel> buhman: Anyway, whatever you do, you need a __gc metamethod. 20181117 02:19:09<+wesdiscordbot> Is there actually and reason why anything from the AI work can't go into 1.14 20181117 02:19:10<+wesdiscordbot> *referenceD issues 20181117 02:19:21< celticminstrel> Which manually calls a destructor. 20181117 02:19:30<+wesdiscordbot> The AI runs on the hosts, so it wouldn't cause OOS 20181117 02:19:36< celticminstrel> ie, p->~window() or p->~unique_ptr() or the like. 20181117 02:20:02< buhman> hmm, is that what UBox does? 20181117 02:20:20<+wesdiscordbot> YWell, that's true, it wouldn't cause OOS. I've always gone by the fact that behavior should be consistent within a stable version. 20181117 02:20:33< celticminstrel> UBox? 20181117 02:20:45<+wesdiscordbot> fact? Maybe I mean "guideline" or something. 20181117 02:21:04< buhman> src/lua/lauxlib.cpp 20181117 02:21:10< celticminstrel> ??? 20181117 02:21:27< celticminstrel> Suddenly you're linking the Lua source code? o.O 20181117 02:21:36<+wesdiscordbot> the trend changed recently a litte 20181117 02:21:38< buhman> no :) 20181117 02:21:40< celticminstrel> You shouldn't worry about anything in src/lua... 20181117 02:21:48< celticminstrel> But you just did? 20181117 02:21:59<+wesdiscordbot> @sevu Yeah, I know, but I am old and change is hard. 😉 20181117 02:22:01< celticminstrel> lauxlib.c is part of the Lua source. 20181117 02:22:11< buhman> oh, right, I forgot you had it in-tree 20181117 02:22:14< celticminstrel> (Wesnoth changes the extensions to cpp so they compile as C++) 20181117 02:22:23<+wesdiscordbot> Seriously though, I don't really mind one way or the other. 20181117 02:22:32< celticminstrel> You want src/scripting 20181117 02:22:52<+wesdiscordbot> If you (pl.) think the AI changes could or should go into 1.14, that's fine by me. 20181117 02:23:17< celticminstrel> I have no idea what any UBox is and I also know that anything in src/lua isn't going to be thinking about destructors. 20181117 02:24:18< celticminstrel> buhman: FTR, there is a lot of code in Wesnoth that passes a C++ class to Lua as a userdata. 20181117 02:24:22< buhman> I don't understand what you mean my "enough memory for the entire dialog"; does that mean the instance needs to be allocated or copied to a lua-managed memory area? 20181117 02:25:04< celticminstrel> I just mean you need to ask Lua to allocate sizeof(gui2::window) bytes as userdata, and construct the dialog in the location it allocates, which is indeed a Lua-managed memory area. 20181117 02:25:21< celticminstrel> The expression new(L) gui2::window(args) does this for you, 20181117 02:25:46<+wesdiscordbot> @sevu Btw, I did test that the ai_algorithm thing works in both MP and SP, and that it works as default and that you can change the AI from the default (without error messages). 20181117 02:26:12< celticminstrel> Provided you include lua_common.hpp 20181117 02:26:25<+wesdiscordbot> Part of my choice of putting it outside the [default_config] was motivated by making that work smoothly without having to change the existing code much. 20181117 02:26:48< celticminstrel> ... 20181117 02:27:29< celticminstrel> Maaaybe I should've made it take L by reference instead, so that it'll actually fail to compile if you forget to include lua_common.hpp? Assuming lua_State isn't a tyepdef of void* or something. 20181117 02:27:34< celticminstrel> ^typedef 20181117 02:27:52< buhman> celticminstrel: what is that syntax feature called? 20181117 02:28:08< buhman> the new(L) magic 20181117 02:28:30<+wesdiscordbot> what I saw was that :droid did lead to an defeat if you use <= 1.14.5 and used the new attribute 20181117 02:28:51< celticminstrel> Pretty sure it's just called a new operator? Maybe placement new. 20181117 02:28:53<+wesdiscordbot> I'm not totally sure though if I had specified an id with it which was not used 20181117 02:29:07< celticminstrel> Though I think placement new is just the one built-in example of it. 20181117 02:29:14<+wesdiscordbot> might be that I used that workaround … in any case an unlikely case 20181117 02:29:33< buhman> https://en.cppreference.com/w/cpp/language/new#Placement_new seems like the thing 20181117 02:29:37< buhman> nice I didn't know you could do that 20181117 02:30:33<+wesdiscordbot> @sevu Yeah, I am currently only concerned with making this work once the commit is in the codebase. I'll leave the owrkarounds to you. 😉 20181117 02:30:40<+wesdiscordbot> But let me know if anything is not working as it should. 20181117 02:31:58<+wesdiscordbot> @sevu With these interlaced discussions it's a bit hard to make sure nothing is missed. Was there anything else I should have replied to? 20181117 02:32:20<+wesdiscordbot> And are there other changes to the AI that you'd like to see? 20181117 02:32:53<+wesdiscordbot> No, nothing missed. But yeah 20181117 02:35:33<+wesdiscordbot> did you see the 2nd [ai] in the core tag? I haven't tested it out, this idea actully originated when I saw I could change the AI settings in era… though, I think unlike scenario it defined a new ai for the dropdown box instead of changeing paramters of the choosen one. 20181117 02:37:53<+wesdiscordbot> not [core] tag … I mean the core add-on 20181117 02:38:27<+wesdiscordbot> @sevu Yes? If I understand you correctly, that's what it's supposed to do. 20181117 02:40:17<+wesdiscordbot> The only bad thing about that is that it probably goes quite unnotived if there is a third option in the drop down menu, so nobody will use the [era] specific AI... while setting things in scenario are always used 20181117 02:41:46<+wesdiscordbot> maybe one could leave it like it is, but add an option so set the default ai for the era 20181117 02:42:53<+wesdiscordbot> (as opossed to not defining a new ai lgorithm but appling the settings to the default/chosen) 20181117 02:43:35<+wesdiscordbot> But honeslty, I think nobody defines an own AI 20181117 02:44:28<+wesdiscordbot> @sevu Yeah, I am probably the only person in Wesnoth who does that ... 20181117 02:44:58<+wesdiscordbot> But I'll have a look into that, to see if there's an easy way to either define a default in [era] or exclude other AIs. 20181117 02:47:13-!- gfgtdf [~Daniel@x4dbaee28.dyn.telefonica.de] has joined #wesnoth-dev 20181117 02:47:57<+wesdiscordbot> In case you're posting anything about AI on the forums… maybe you can bring up the oportunity to thes the Experimental AI with the core 20181117 02:48:33<+wesdiscordbot> Not planning to at the moment, but I'll keep it in mind. 20181117 02:48:51< gfgtdf> sevu: I don't know which ai changes you mean in particular but ai changed if they have a huge impact on balance/difficulty of campaigns are often unsuitable for stable becasue they will change the campaigns difficulty and in the worst case make it unwinnable. 20181117 02:49:19< gfgtdf> celticminstrel: you had a question ? 20181117 02:49:34< celticminstrel> Yes. 20181117 02:49:38< mattsc> gfgtdf: we’re not talking about those types of AI changes at the moment 20181117 02:50:03< celticminstrel> Why did you feel the need for a separate mapgen_gamemap class? 20181117 02:50:21< mattsc> As in, the default AI is not changing at all. Just the options to use non-default ones are being enhanced. 20181117 02:51:11< celticminstrel> "the oportunity to thes the Experimental AI " --> ??? 20181117 02:51:25< mattsc> s/thes/test, I presume 20181117 02:51:30< gfgtdf> you mea as opposed to the original c++ gamemap class or your lau map class ? 20181117 02:51:42< celticminstrel> The original C++ gamemap class. 20181117 02:52:22< celticminstrel> Your explanatory comment doesn't seem to make sense. 20181117 02:53:59<+wesdiscordbot> Btw, I the [ais] and [default_config] tags aren't in the list of the ReferenceWML page… are they somewhere documented? 20181117 02:54:01< gfgtdf> mostly for technical reasons iirc, the gamemap class uses terrain_data_cache which we don't have there. Also the set_terrain() bheves slightly differntly as the lua gamelap class does not check whetehr the terain is valid 20181117 02:54:48< gfgtdf> it's likeley to be possible to merge some code like having a common baseclass or making the gamemap a subclass of the map clas that the lua map uses. 20181117 02:54:50< celticminstrel> So how do you propose to use the actual C++ gamemap class here when mapgen has a separate class 20181117 02:54:55< celticminstrel> Yeah okay. 20181117 02:55:05< celticminstrel> Hmm. 20181117 02:55:25< celticminstrel> What is the terrain_data_cache and why doesn't mapgen have it? 20181117 02:57:17<+wesdiscordbot> @sevu I don't think they are documented; presumably because until recently they really weren't modifiable other than in a hack-into-default files way. 20181117 02:58:08< gfgtdf> it contains information about the terrain liek which terrain is a village which terrains are 'known', i think it could be possible to pass it around to that to make is accessible to that. 20181117 02:58:53< celticminstrel> Does the main gamemap have a rough equivalent of simplemerge? 20181117 03:00:57-!- irker078 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181117 03:00:57< irker078> wesnoth/wesnoth:master mattsc a29e25ebae Command-line game: get default ai_algori AppVeyor: All builds passed 20181117 03:01:42< mattsc> irker078: you are slow 20181117 03:02:48< celticminstrel> Also, not strictly related to the mapgen, but how is luaW_tableget different from lua_getfield? 20181117 03:04:29< gfgtdf> celticminstrel: no it doesn't, iirc, but that just a 6 lien function that can be added 20181117 03:06:31< gfgtdf> celticminstrel, i think te main differnce is that it returns a bool so that it supposed to be used as `if(luaW_tableget(L, -1, "old")) {` if you don't want to execute code if the attribute is not set 20181117 03:07:30< gfgtdf> celticminstrel: for most of the usecases you can use luaW_table_get_def now though actually. 20181117 03:08:13< gfgtdf> celticminstrel: which is easier to use since it jsut returns the value instead of pushing it on the stack. 20181117 03:08:29< celticminstrel> ...that's even worse. o.O 20181117 03:08:45< celticminstrel> Amyway, does that mean it's fine to remove luaW_tableget? 20181117 03:08:56< celticminstrel> ^anyway 20181117 03:09:10-!- sevu [~sevu@p5485415E.dip0.t-ipconnect.de] has joined #wesnoth-dev 20181117 03:09:21-!- sevu [~sevu@p5485415E.dip0.t-ipconnect.de] has quit [Remote host closed the connection] 20181117 03:09:22< gfgtdf> not sure, look at the parts that use it, if they can be converted to luaW_table_get_def, then probably yes. 20181117 03:09:39< celticminstrel> Or if they can be converted to lua_getfield? 20181117 03:13:13< gfgtdf> i still think luaW_tableget is easier to use than lua_getfield for those cases but i don't care that much. 20181117 03:25:34-!- gfgtdf [~Daniel@x4dbaee28.dyn.telefonica.de] has quit [Ping timeout: 272 seconds] 20181117 03:25:48-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has joined #wesnoth-dev 20181117 03:51:27<+wesdiscordbot> Feedback for the game tutorial (only Wesnoth 1.1.2 and later.) 20181117 03:52:03<+wesdiscordbot> It really mentions 1.1.2 in the forum. Does someone mind changing it (it's int he Campaign feedback Section) 20181117 03:55:11< celticminstrel> o.o 20181117 04:47:16-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has quit [Read error: Connection reset by peer] 20181117 04:49:06<+wesdiscordbot> gfgtdf 20181117 04:49:06<+wesdiscordbot> https://cdn.discordapp.com/attachments/259976436490829825/513213966936309771/robots.png 20181117 05:31:48< celticminstrel> 'kay, got the code compiling with the mapgen_gamemap class removed, next I need to see if it still works. 20181117 05:32:37< celticminstrel> I think gfgtdf said he had a test somewhere? 20181117 05:34:21< celticminstrel> I see, in data/multiplayer/scenarios... why the heck is it in such a weird place though. 20181117 05:34:34-!- celticminstrel is now known as celmin|sleep 20181117 06:01:06-!- irker078 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181117 07:43:41-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20181117 08:27:08<+wesdiscordbot> @sevu done 20181117 11:10:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20181117 12:35:46-!- irker460 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181117 12:35:46< irker460> wesnoth: gfgtdf wesnoth:1.14 ffe58acad75b / src/menu_events.cpp: show more scenarios in choose_level debug command https://github.com/wesnoth/wesnoth/commit/ffe58acad75bf18ae7bc5b002144d34c1ea3af33 20181117 12:35:46< irker460> wesnoth: gfgtdf wesnoth:1.14 b51c7dfa765e / / (114 files in 50 dirs): Merge branch '1.14' of https://github.com/wesnoth/wesnoth into 1.14 https://github.com/wesnoth/wesnoth/commit/b51c7dfa765e2587a5b174fe40678088a098eec7 20181117 12:35:47< irker460> wesnoth: gfgtdf wesnoth:1.14 2de6b80218ad / src/server/ (game.cpp game.hpp): wesnothd: add game::get_player_versions() https://github.com/wesnoth/wesnoth/commit/2de6b80218add7033db4376882ff2fd1417284b7 20181117 12:35:48< irker460> wesnoth: gfgtdf wesnoth:1.14 c445828e6206 / src/server/game.cpp: don't send whiteboard data to older clients https://github.com/wesnoth/wesnoth/commit/c445828e62067749149cc410a7da6c914a714d44 20181117 12:40:41-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has joined #wesnoth-dev 20181117 12:42:52< gfgtdf> Soliton, shadowm: I made a server commit ^ to fix an oos error, please rebuild the server at some time. 20181117 14:31:33< irker460> wesnoth/wesnoth:1.14 Steve Cotton 7c29973b95 Tutorial S2: "choose who gets the 16xp" AppVeyor: All builds passed 20181117 14:58:40-!- celmin|sleep is now known as celticminstrel 20181117 15:27:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20181117 15:56:24-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has quit [Remote host closed the connection] 20181117 15:56:54-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has joined #wesnoth-dev 20181117 16:01:24-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has quit [Remote host closed the connection] 20181117 16:52:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20181117 16:55:50-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc] 20181117 16:57:03-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20181117 16:57:15-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Client Quit] 20181117 17:11:49< celticminstrel> So my agenda for this weekend is: merge the liminal PR. 20181117 17:12:36< celticminstrel> Can I let someone else take care of adjusting Dunefolk damage to compensate as needed? 20181117 17:15:06<+wesdiscordbot> liminal is going to change? 20181117 17:26:04-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has joined #wesnoth-dev 20181117 17:32:08-!- irker460 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181117 17:44:46< celticminstrel> Invert. 20181117 17:44:53< celticminstrel> Roughly. 20181117 17:45:21< celticminstrel> The overall effect under the default schedule should be the same, but the overall effect under other schedules might differ. 20181117 17:56:09<+wesdiscordbot> I see. I like. 20181117 18:02:20<+wesdiscordbot> celticminstrel: isn't their damage changing as part of the same PR? 20181117 18:02:25< celticminstrel> Ohh, so it is. 20181117 18:02:32< celticminstrel> It'd been so long since I looked at it that I forgot. 20181117 18:05:27-!- irker631 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181117 18:05:27< irker631> wesnoth: fsd wesnoth:master 40483874eae7 / src/ (5 files in 2 dirs): change liminal implementation, from a night and day malus to a twilight bonus https://github.com/wesnoth/wesnoth/commit/40483874eae762c4f4eb10c76f60c40b5603f245 20181117 18:05:27< irker631> wesnoth: fsd wesnoth:master 78c71dc7531c / data/ (core/help.cfg english.cfg): update help to reflect the liminal change https://github.com/wesnoth/wesnoth/commit/78c71dc7531cee2fb02934db5eb6e2e3be934ea2 20181117 18:05:28< irker631> wesnoth: fsd wesnoth:master 52cee19e3655 / data/core/units/dunefolk/ (10 files): adjust liminal unit stats after liminal change from malus to bonus https://github.com/wesnoth/wesnoth/commit/52cee19e36553ee92c1aa812ee7daa508010f56c 20181117 18:05:29< irker631> wesnoth: fsd wesnoth:master ec1c7cb2d8d6 / src/tod_manager.cpp: use has_attribute(liminal_bonus) as suggested https://github.com/wesnoth/wesnoth/commit/ec1c7cb2d8d6d9b049c1a28681d8bac96c73616f 20181117 18:05:30< irker631> wesnoth: Celtic Minstrel wesnoth:master 7b358440f6c6 / / (17 files in 5 dirs): Merge pull request #2664 from holius-fr/liminal_bonus https://github.com/wesnoth/wesnoth/commit/7b358440f6c6e5083185b16629c8b61ae54ed511 20181117 18:12:14<+wesdiscordbot> @shadowm Can master's units database be updated as well? It looks like it hasn't been run since April anyway. 20181117 18:12:55< celticminstrel> Maybe ping elias on that, I think there was some reason it isn't being run... 20181117 18:13:13< celticminstrel> Like a bug that needs to be fixed or something? I don't remember. 20181117 18:15:41< irker631> wesnoth/wesnoth:master Celtic Minstrel 7b358440f6 Merge pull request #2664 from holius-fr/ AppVeyor: vs2015/Release Failed 20181117 18:15:42< irker631> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20376139 20181117 18:20:06-!- travis-ci [~travis-ci@ec2-23-22-15-60.compute-1.amazonaws.com] has joined #wesnoth-dev 20181117 18:20:07< travis-ci> wesnoth/wesnoth#20221 (master - 7b35844 : Celtic Minstrel): The build was broken. 20181117 18:20:07< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/456409436 20181117 18:20:07-!- travis-ci [~travis-ci@ec2-23-22-15-60.compute-1.amazonaws.com] has left #wesnoth-dev [] 20181117 18:20:28-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 272 seconds] 20181117 18:30:10< elias> I thought shadowm fixed some unit database issue... but maybe that was indeed in april 20181117 18:38:02-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20181117 19:18:20< irker631> wesnoth/wesnoth:1.14 gfgtdf c445828e62 don't send whiteboard data to older clie AppVeyor: All builds passed 20181117 19:37:08-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev 20181117 19:52:46-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 260 seconds] 20181117 19:54:32-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev 20181117 20:01:38-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20181117 20:32:51< gfgtdf> celticminstrel: you can access the test from the mapeditor->generatemap aswell. 20181117 20:37:29-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc] 20181117 20:54:48<+wesdiscordbot> @Pentarctagon Why would we need to update the master unit database when there isn't a 1.15.x add-ons server yet? 20181117 20:55:01<+wesdiscordbot> (And that's why I disabled the cronjob.) 20181117 21:01:57<+wesdiscordbot> gfgtdf: How urgent is the server rebuilt supposed to be? I'm asking because I got a GH mention in my emails 20181117 21:02:16<+wesdiscordbot> It'd be nice if you could say this in general since I obviously don't want to rebuild and restart the server during peak hours 20181117 21:02:29<+wesdiscordbot> s/in general/in the future/ 20181117 21:02:58<+wesdiscordbot> (I can do it if it's actually "people will riot if we don't fix this" urgent obviously) 20181117 21:03:55< gfgtdf> it fixes a oos when using planning mode in coop games, i don't think people will riot. 20181117 21:05:41< gfgtdf> or to be more specific when using planned recruits in coop games. 20181117 21:06:30< gfgtdf> It'd be nice to have it done in like the next 12 hours. 20181117 21:06:37< gfgtdf> i don't know when the peak hours are 20181117 21:07:09< gfgtdf> It be also nice if you would put a comment in that gh thread when you have rebuild it. 20181117 21:09:01<+wesdiscordbot> During weekends it's basically the whole day until around 10 pm PST 20181117 21:09:16<+wesdiscordbot> (6 am UTC) 20181117 21:09:55<+wesdiscordbot> I'll rebuild it in the meantime but probably won't restart it until around 3 am UTC 20181117 21:11:34< gfgtdf> ok 20181117 21:12:18< gfgtdf> that commend in the gitoub thread when was about when you ahve restarted it obviously. 20181117 21:12:47< gfgtdf> shadowm: i thought the server rebuilds automaticialyl every hour and you just ahev to restart it? At least that iirc what you told me at some time. 20181117 21:17:11<+wesdiscordbot> No, it doesn't 20181117 21:17:28<+wesdiscordbot> What it does automatically every hour is to pull from upstream into the local Git repository 20181117 21:39:56-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20181117 21:47:14-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20181117 21:52:00<+wesdiscordbot> All: I assume that not everybody here reads the Github issue comments. I made some suggestions with changes to the default AI for 1.15 that will definitely affect balance. If you have opinions on that, please reply in the issue tracker. https://github.com/wesnoth/wesnoth/issues/3450#issuecomment-439645681 20181117 22:18:31-!- irker631 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181117 22:36:11-!- gfgtdf [~Daniel@x4d084012.dyn.telefonica.de] has quit [Quit: Leaving] 20181117 22:48:35<+wesdiscordbot> @shadowm It would just be convenient to be able to link to updated unit stats post https://github.com/wesnoth/wesnoth/pull/2664 20181117 22:49:01<+wesdiscordbot> But there's no 1.15.x add-ons server yet 20181117 22:49:36<+wesdiscordbot> so? 20181117 22:50:27<+wesdiscordbot> this is related to the dunefolk balancing thread(s), not add-ons 20181117 22:57:57<+wesdiscordbot> I know 20181117 22:58:05<+wesdiscordbot> But the scripts want to see an add-ons server 20181117 22:58:26<+wesdiscordbot> And I cba to change that because I didn't write them 20181117 22:59:49<+wesdiscordbot> oh 20181117 23:02:39<+wesdiscordbot> What does it take to get the add-ons server for 1.15 up? We'd need that for the eventual 1.15.0 release as well. 20181117 23:12:24-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 268 seconds] 20181117 23:15:11<+wesdiscordbot> Just a few things 20181117 23:15:29<+wesdiscordbot> I mean I can do it later I guess, it's actually very easy 20181117 23:15:47-!- irker209 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181117 23:15:47< irker209> wesnoth/wesnoth:master Celtic Minstrel 7b358440f6 Merge pull request #2664 from holius-fr/ AppVeyor: 2/4 builds failed 20181117 23:15:47< irker209> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20376139 20181117 23:15:47< irker209> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/builds/20376140 20181117 23:15:52<+wesdiscordbot> I guess it's been a while since the 1.14.0 release so it's kind of justified at this point 20181117 23:24:12<+wesdiscordbot> Alright. It's not urgent, obviously, but whenever you're able to get around to it would be great. --- Log closed Sun Nov 18 00:00:38 2018