--- Log opened Fri Nov 16 00:00:36 2018 20181116 00:01:09<+wesdiscordbot> @josteph about your comment on the "advanced strategies" in the tutorial, you could look at @Inky 's tactics puzzles on the addon server 20181116 00:01:51<+wesdiscordbot> she's already done a few basic strategy-type things, maybe that would fit with what you're thinking of 20181116 00:02:57-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20181116 00:31:54< irker735> wesnoth/wesnoth:master mattsc 233e30df15 Another fix to AI aspect unit tests AppVeyor: vs2015/Release Failed 20181116 00:31:55< irker735> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20339008 20181116 00:39:36<+wesdiscordbot> @Yumi I I thought the same, in fact I'd pm'd Inky already. šŸ˜ƒ 20181116 00:40:50<+wesdiscordbot> Ravana_, what's that about the bot? What's not working ? 20181116 00:52:51<+wesdiscordbot> :p 20181116 01:00:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20181116 01:19:46< irker735> wesnoth/wesnoth:master newfrenchy83 98e1c984e6 fix indentation AppVeyor: 1/2 builds failed 20181116 01:19:47< irker735> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/builds/20334112 20181116 01:35:29< irker735> wesnoth/wesnoth:master mattsc b1ad86aeba Unit tests: change attack_depth to leade AppVeyor: vs2017/Release Failed 20181116 01:35:30< irker735> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/builds/20341564 20181116 01:58:42-!- celmin|away is now known as celticminstrel 20181116 01:58:59< celticminstrel> Does anyone know why --wconsole would interfere with output redirection? 20181116 01:59:43< celticminstrel> Also, since no-one commented on lua_reorg I may merge it now instead of in two days. 20181116 01:59:57<+wesdiscordbot> Because output redirection is set up during process initialization 20181116 02:00:07< celticminstrel> Meaning...? 20181116 02:00:16<+wesdiscordbot> Before the process' entry point is executed 20181116 02:00:27<+wesdiscordbot> --wconsole sets up the console and the stdout and stderr long after that 20181116 02:00:31< celticminstrel> Ah. 20181116 02:00:46< celticminstrel> I guess there's no way for it to tell that output has been redirected, huh. 20181116 02:01:46<+wesdiscordbot> I'm pretty sure the file descriptors in question are pointing to \Device\Null either way unless the process image is flagged as a console subsystem app 20181116 02:01:55< celticminstrel> I see. 20181116 02:06:58< irker735> wesnoth: Celtic Minstrel wesnoth:master e3e80b4a9905 / data/lua/ (core.lua helper.lua wml/items.lua): Lua API reorganization: interface module https://github.com/wesnoth/wesnoth/commit/e3e80b4a99058e6032beed2d840f08dd84d0ac42 20181116 02:07:00< irker735> wesnoth: Celtic Minstrel wesnoth:master 55a6c16810a0 / data/lua/ (core.lua helper.lua wml/items.lua): Lua API reorganization: interface module deprecations https://github.com/wesnoth/wesnoth/commit/55a6c16810a03f51f96f1f807e66a55fcbe50daa 20181116 02:07:02< irker735> wesnoth: Celtic Minstrel wesnoth:master 3ec9ce4b60e3 / data/lua/ (core.lua helper.lua): Lua API reorganization: GUI module https://github.com/wesnoth/wesnoth/commit/3ec9ce4b60e3b0678daac35f71ffac5c27b4533c 20181116 02:07:04< irker735> wesnoth: Celtic Minstrel wesnoth:master 3e79b3bde841 / data/lua/ (core.lua helper.lua): Lua API reorganization: GUI module deprecations https://github.com/wesnoth/wesnoth/commit/3e79b3bde841383140d8cae7e82c6130ba2e4c7d 20181116 02:07:06< irker735> wesnoth: Celtic Minstrel wesnoth:master 7c3fd9a612b0 / data/lua/ (core.lua helper.lua): Lua API reorganization: units module https://github.com/wesnoth/wesnoth/commit/7c3fd9a612b03d7f7f44d3405bdf7202e75c0539 20181116 02:07:08< irker735> wesnoth: Celtic Minstrel wesnoth:master 8bcf749666af / data/lua/ (core.lua helper.lua): Lua API reorganization: units module deprecations https://github.com/wesnoth/wesnoth/commit/8bcf749666af8741d130be5bca9a8d724583c3d3 20181116 02:07:10< irker735> wesnoth: Celtic Minstrel wesnoth:master 6d7736e6e61c / data/lua/ (core.lua helper.lua): Deprecate helper.wml_error and move it to wml.error https://github.com/wesnoth/wesnoth/commit/6d7736e6e61c231e90f02ec09bf1c303e27e3bf9 20181116 02:17:12< irker735> wesnoth: Celtic Minstrel wesnoth:master 866420adf85e / data/ (core/macros/utils.cfg scenario-test.cfg): Redesign force cth macro https://github.com/wesnoth/wesnoth/commit/866420adf85ec1da02f565f2efd1933bf27419c4 20181116 02:22:51< celticminstrel> So, everyone knows that math.random is not safe for use in MP and all that. 20181116 02:23:08< celticminstrel> Is there any reason whatsoever that we can't simply replace math.random with wesnoth.random? 20181116 02:23:23< celticminstrel> ie, as if assigning math.random = wesnoth.random 20181116 02:23:57< celticminstrel> Some people seem to have objections to adding anything to the core Lua modules, but this isn't even a case of that, really? The new function pretty much does the same thing as the old, right? 20181116 02:36:24< celticminstrel> ...gfgtdf added a map Lua object but it's useless for in-game, whyyyyy. 20181116 02:40:04<+wesdiscordbot> then remove it? 20181116 02:40:34< celticminstrel> I think you missed the "for in-game" part. It's useful for generating maps, I'm just griping about how he didn't bother to make it work in-game too. 20181116 02:41:03< celticminstrel> That said, I might actually remove the C++ class underlying the Lua map API. 20181116 02:41:24< celticminstrel> But I should talk to gfgtdf first to try and understand why he thought it was necesary to add a second map class. 20181116 02:41:45< irker735> wesnoth/wesnoth:master newfrenchy83 8bef5dce57 Update utils.cfg AppVeyor: All builds passed 20181116 02:43:51< celticminstrel> @Vultraz: If you saw wesnoth.map[5][6] what location would you expect that to return? 20181116 02:44:02<+wesdiscordbot> 5,6 20181116 02:44:10< celticminstrel> So x=5 y=6 huh. 20181116 02:44:17<+wesdiscordbot> I mean 20181116 02:44:18<+wesdiscordbot> yes? 20181116 02:44:22<+wesdiscordbot> it's intuitive 20181116 02:44:26<+wesdiscordbot> and lua is base-1 anyway 20181116 02:44:27< celticminstrel> IOW, column-major access. 20181116 02:44:38< celticminstrel> Oh, it wasn't the base I was worried about. 20181116 02:44:41< celticminstrel> It was the order. 20181116 02:44:55<+wesdiscordbot> coordinates are never y,x 20181116 02:45:14< celticminstrel> Tell that to everyone who uses [y][x] for their 2D arrays. 20181116 02:45:37< celticminstrel> FTR, I imagine the top left might be [0][0] or even though Lua is 1-based, because borders. 20181116 02:45:42< celticminstrel> ^or so 20181116 02:45:54<+wesdiscordbot> yes 20181116 02:47:53<+wesdiscordbot> anyway, I'm not going to schedule 1.15.0 yet since there's nothing interesting. 20181116 02:48:02<+wesdiscordbot> for players 20181116 02:48:10< celticminstrel> Maybe after I merge the mapgen PR then, huh? 20181116 02:48:15< celticminstrel> Since that adds a new random map. 20181116 02:48:23<+wesdiscordbot> perrrrrhaps 20181116 02:48:33< celticminstrel> Not expecting that to happen this month though. 20181116 02:49:07<+wesdiscordbot> I'm also debating whether it's worth sharing one or two screenshots of series 2 and writing about it more publicly. Promising too much early is often a bad idea, but we are likely going to have to launch a patreon... 20181116 02:49:24<+wesdiscordbot> or rather, the patreon we've been discussing launching for years and never got around to it 20181116 02:49:25< celticminstrel> Getting a little tempted to just revert all gfgtdf's mapgen work and start over from scratch, but... I'll try yo hold back until I get a chance to talk to him. 20181116 02:49:30< celticminstrel> ^try to 20181116 02:49:59< celticminstrel> If you need a patreon for series 2 it doesn't really inspire confidence in me. 20181116 02:50:12<+wesdiscordbot> Nothing inspires confidence in you 20181116 02:50:18<+wesdiscordbot> WE'VE BEEN DISCUSSING THE IDEA OF A PATREON BEFORE THE IDEA OF SERIES 2 CAME UOP 20181116 02:50:21<+wesdiscordbot> Oh my god 20181116 02:50:22< celticminstrel> You'd be surprised. 20181116 02:50:27<+wesdiscordbot> I just said that. 20181116 02:50:29< celticminstrel> The point is "need". 20181116 02:50:34<+wesdiscordbot> " the patreon we've been discussing launching for years and never got around to it" 20181116 02:50:39<+wesdiscordbot> You came up with "need" 20181116 02:50:41< celticminstrel> Or "have to" to use his exact phrasing. 20181116 02:50:49<+wesdiscordbot> Vultraz didn't say anything about "need" 20181116 02:51:00< celticminstrel> I did not come up with "need", Vultraz directly said "have to launch" which is the same as "need to launch". 20181116 02:51:14<+wesdiscordbot> Do you realize that Wesnoth's only income source is iOS? 20181116 02:51:18< celticminstrel> Yes. 20181116 02:51:23<+wesdiscordbot> Do you realize how difficult it is to commission art like this? 20181116 02:51:31<+wesdiscordbot> I said likely. A big project like series 2 is an optimal time to launch a Patreon, 20181116 02:51:49<+wesdiscordbot> Because its the kind of big thing players like funding 20181116 02:51:55< celticminstrel> True, you did say "likely". 20181116 02:51:59<+wesdiscordbot> As opposed to "please let us pay our server bills" 20181116 02:52:14<+wesdiscordbot> (not that we're struggling for money in that department) 20181116 02:52:16<+wesdiscordbot> (to be clear) 20181116 02:52:22<+wesdiscordbot> (we have more than enough for the hardware) 20181116 02:52:36< celticminstrel> I don't think you should start soliciting Patreon donations for series 2 until you're at a point where it's proven to be definitely viable. 20181116 02:52:38<+wesdiscordbot> but we;ve really had nothing to inspire people to donate 20181116 02:53:09<+wesdiscordbot> It's not for series 2, it's for the Battle for Wesnoth Project as a whole 20181116 02:53:15< celticminstrel> Then that's fine. 20181116 02:53:47<+wesdiscordbot> Yes. 20181116 02:54:00<+wesdiscordbot> That was always the intention 20181116 02:54:17<+wesdiscordbot> Hell, we were on the verge of launching it a few months ago 20181116 02:54:21< celticminstrel> Still, advertising it as part of soliciting donations seems poor form when you're still not 90% sure it's actually going to go forward. 20181116 02:54:27< celticminstrel> "it" -> series 2 20181116 02:55:01< celticminstrel> Does gfgtdf actually come on IRC anymore? Should I ping him on GitHub or something? 20181116 02:55:11<+wesdiscordbot> You really have no confidence, do you. >_> A project succeeds if people work on it. If you decide it's not worth working on, then of course it will fail. 20181116 02:55:29<+wesdiscordbot> Projects don't just "happen" 20181116 02:55:38< celticminstrel> Projects don't succeed just because people work on them. 20181116 02:55:47< celticminstrel> That's important, sure, but it's not the only factor. 20181116 02:55:53<+wesdiscordbot> You can't wait around staring at the sky pondering if the amorphous idea of a project will come falling to earth. 20181116 02:56:16<+wesdiscordbot> Of course things people work on can fail 20181116 02:56:21<+wesdiscordbot> Terraria: Otherworlds failed 20181116 02:56:23<+wesdiscordbot> for example 20181116 02:56:40<+wesdiscordbot> but they actually committed to trying to deliver it. You seem content to not even try that. 20181116 02:57:23<+wesdiscordbot> Your logic RE funding is tautological. Wait to see if it works before seeking funding to help realize it 20181116 02:58:11< celticminstrel> Maybe that's because the project is ill-advised from the start. 20181116 02:58:30<+wesdiscordbot> le deep sigh 20181116 02:58:33<+wesdiscordbot> Can you take your thinly-veiled hostility towards the work on series 2 somewhere where we can't see it? 20181116 02:59:25<+wesdiscordbot> I'm not even involved because of personal reasons but I don't use that as an excuse to go around GOING OUT OF MY WAY to try to demotivate people working on it 20181116 02:59:36<+wesdiscordbot> That's just shitty 20181116 03:11:25<+wesdiscordbot> celmin, about math.random, overriding stdlib functions is usually a bad idea. Is it possible to add a linter warning or something to catch wrong uses of it ? 20181116 03:11:45< celticminstrel> Why is it a bad idea? Seriously, why? 20181116 03:11:57< celticminstrel> Especially when the new function pretty much does the exact same thing? 20181116 03:12:36<+wesdiscordbot> Because someone might have a legitimate reason to use math.random 20181116 03:12:39< celticminstrel> Plus, we already override print(), load(), dofile(), and require(), though the latter two under the wesnoth namespace. 20181116 03:12:47<+wesdiscordbot> ... and would run into the differences 20181116 03:13:00< celticminstrel> Is there any legitimate reason to use math.random that would not be satisfied by using wesnoth.random? 20181116 03:13:09<+wesdiscordbot> "override under another namespace" is different, it's opt-in 20181116 03:13:24< celticminstrel> It's not opt-in. 20181116 03:13:28< celticminstrel> The global versions do not exist. 20181116 03:13:38<+wesdiscordbot> that counts 20181116 03:13:54<+wesdiscordbot> if someone just types 'dofile', they aren't going to get the wesnoth version when they expected the lua version 20181116 03:14:14<+wesdiscordbot> it's fine if they get an error - better an error than silently different behavior 20181116 03:14:24-!- travis-ci [~travis-ci@ec2-54-162-92-105.compute-1.amazonaws.com] has joined #wesnoth-dev 20181116 03:14:25< travis-ci> wesnoth/wesnoth#20210 (master - 6d7736e : Celtic Minstrel): The build has errored. 20181116 03:14:25< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/455789392 20181116 03:14:25-!- travis-ci [~travis-ci@ec2-54-162-92-105.compute-1.amazonaws.com] has left #wesnoth-dev [] 20181116 03:14:41<+wesdiscordbot> math.random might be faster and one could use it for local UI things? Or math.random might be extended in the next lua? 20181116 03:15:12<+wesdiscordbot> I should say, I'm not overly familiar with lua, but I assume it's similar to python in this regard 20181116 03:15:45<+wesdiscordbot> Math.random() potentially being extended in future versions of Lua is irrelevant. We decide when developers get major Lua upgrades. 20181116 03:17:19<+wesdiscordbot> Even so, if people write code that calls math.random expecting wesnoth.random semantics, and then we upgrade lua some day... 20181116 03:17:30< celticminstrel> It's true that wesnoth.require and global require take different input, I guess. And I suppose that's technically true with dofile too, given the virtual WML filesystem... 20181116 03:17:41<+wesdiscordbot> And we could upgrade our own implementation at the same time. 20181116 03:18:11<+wesdiscordbot> Right. But if people wrote wesnoth.random to begin with, we wouldn't have to extend our implementation. 20181116 03:19:02< celticminstrel> I suppose there might be one possible problem with moving wesnoth.random to math.random. 20181116 03:19:03<+wesdiscordbot> Or, I don't know the signature of this, can we write wesnoth.random in a generic way, that doesn't depend on the signature of math.random ? 20181116 03:19:08< celticminstrel> What do you do with math.randomseed? 20181116 03:19:25< celticminstrel> Pretty sure math.random and wesnoth.random have the exact same signature BTW. 20181116 03:19:27<+wesdiscordbot> That sounds like a very weak advantage, compared with developers having to write less idiomatic Lua, and needing to spend additional work to port third-party libraries to Wesnoth's implementation of Lua. 20181116 03:19:46-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20181116 03:20:00< celticminstrel> Not sure if the seed issue is a problem with moving wesnoth.random to math.random. 20181116 03:20:24< celticminstrel> Is it even possible to safely reseed the synced random generator? 20181116 03:20:29<+wesdiscordbot> So we want all math.random calls to be implicitly synced ? 20181116 03:21:18< celticminstrel> I don't actually know what this syncing really does... 20181116 03:21:24<+wesdiscordbot> What if someone wanted to do a non-synced roll (for UI purposes, say) 20181116 03:22:11< celticminstrel> Anyway, it seems like there might be real reasons why we can't just move wesnoth.random to math.random. 20181116 03:22:29< celticminstrel> I think doing so would be nice, but... 20181116 03:23:18<+wesdiscordbot> @sevu https://github.com/wesnoth/wesnoth/issues/3724 is it something I did? 20181116 03:24:31<+wesdiscordbot> Might be that you patched that 20181116 03:25:15<+wesdiscordbot> But it's not regression from that, or I think it's from somewhere in 1.13 20181116 03:27:11<+wesdiscordbot> Yeah, I was curious if I'd regressed it 20181116 03:27:50<+wesdiscordbot> How do you reproduce it? I can't reproduce it in the test scenario https://gist.github.com/jostephd/ec2a8c24ff68331b1412231e4be74936 20181116 03:29:14<+wesdiscordbot> Hmmmā€¦ I don't have a save from that anymore 20181116 03:30:18-!- sevu [~sevu@p54855FAA.dip0.t-ipconnect.de] has quit [Remote host closed the connection] 20181116 03:30:51<+wesdiscordbot> by the way, it's funny how the minimap shows the terrain even when the map is blacked out 20181116 03:33:09<+wesdiscordbot> I don't even know how it was blacked out, it was not permanently, just for that message 20181116 03:36:12<+wesdiscordbot> what's more, [message] doesn't result in yellow hexes 20181116 03:36:37<+wesdiscordbot> so something is missing 20181116 03:43:20<+wesdiscordbot> Then I best ask the developerā€¦ 20181116 03:47:46< irker735> wesnoth: Jyrki Vesterinen wesnoth:master 68f3a8e84659 / data/lua/ (core.lua wml/items.lua): Lua fixes https://github.com/wesnoth/wesnoth/commit/68f3a8e846599f2ca0636f5570fd26d279590a49 20181116 03:52:55< celticminstrel> :| 20181116 03:53:02< celticminstrel> ...how did that "hex" disappear, anyway? 20181116 03:54:15< celticminstrel> @jyrkive - I don't think those last three lines are necessary - the idea is that wesnoth.interface.* are the canonical names while methods.* are the deprecated names. 20181116 03:54:43<+wesdiscordbot> I reordered the lines. 20181116 03:55:04< celticminstrel> Huh? 20181116 03:55:10<+wesdiscordbot> The problem with older code was that it attempted to use the "methods" table before defining it. 20181116 03:55:12< celticminstrel> This is in items.lua FTR. 20181116 03:55:20< celticminstrel> Ohh. 20181116 03:55:44< celticminstrel> That was probably a relic of the previous commit. Those three lines should've been removed when I deprecated them. :/ 20181116 03:56:19< celticminstrel> Or to put it another way, the local methods = { ... } was meant to replace those three lines. 20181116 04:11:04-!- travis-ci [~travis-ci@ec2-54-81-66-218.compute-1.amazonaws.com] has joined #wesnoth-dev 20181116 04:11:05< travis-ci> wesnoth/wesnoth#20211 (master - 866420a : Celtic Minstrel): The build has errored. 20181116 04:11:05< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/455791414 20181116 04:11:05-!- travis-ci [~travis-ci@ec2-54-81-66-218.compute-1.amazonaws.com] has left #wesnoth-dev [] 20181116 04:12:59< irker735> wesnoth: Celtic Minstrel wesnoth:master 2f9bca43afa8 / data/lua/wml/items.lua: Fix erroneous deprecation https://github.com/wesnoth/wesnoth/commit/2f9bca43afa8fbe98ddf8ca219ebe6b1ed65dfff 20181116 04:16:03< mattsc> celticminstrel: Iā€™ll directly reply to your comments some other time, but as for the unit tests: the numbers needed to be changed because the default for attack_depth and leader_value is different. 20181116 04:16:16< celticminstrel> Ah, I see. 20181116 04:16:50< mattsc> ā€¦ and I donā€™t really know what the ā€œfriendā€ thing means in a C++ context. But I am sure Google can tell me. 20181116 04:17:03< celticminstrel> "friend" is a C++ keyword. 20181116 04:17:26< celticminstrel> It declares a function or class as having access to private data and functions in the class it appears in. 20181116 04:17:39< mattsc> Oh, I see. 20181116 04:18:14< celticminstrel> It sounded like the TODO was complaining that a friend declaration was needed for that line to work, so if that's the case and there is such a friend declaration and it's no longer needed, then it should be removed. 20181116 04:18:57< mattsc> Agreed that thatā€™s how it sounded. Iā€™ll check that out. 20181116 04:20:10< mattsc> And the schema thing was a mistake, donā€™t know why I did that like that. 20181116 04:20:40< mattsc> Probably because I was not paying attention after just having removed a gazillion attack_depth lines from mainline scenarios. 20181116 04:29:14-!- travis-ci [~travis-ci@ec2-54-163-94-123.compute-1.amazonaws.com] has joined #wesnoth-dev 20181116 04:29:15< travis-ci> wesnoth/wesnoth#20212 (master - 68f3a8e : Jyrki Vesterinen): The build passed. 20181116 04:29:15< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/455809616 20181116 04:29:15-!- travis-ci [~travis-ci@ec2-54-163-94-123.compute-1.amazonaws.com] has left #wesnoth-dev [] 20181116 04:31:43< irker735> wesnoth/wesnoth:master mattsc 233e30df15 Another fix to AI aspect unit tests AppVeyor: 1/4 builds failed 20181116 04:31:44< irker735> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20339008 20181116 04:45:19-!- celticminstrel is now known as celmin|sleep 20181116 04:48:22-!- travis-ci [~travis-ci@ec2-54-163-94-123.compute-1.amazonaws.com] has joined #wesnoth-dev 20181116 04:48:23< travis-ci> wesnoth/wesnoth#20213 (master - 2f9bca4 : Celtic Minstrel): The build passed. 20181116 04:48:23< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/455813838 20181116 04:48:23-!- travis-ci [~travis-ci@ec2-54-163-94-123.compute-1.amazonaws.com] has left #wesnoth-dev [] 20181116 04:50:56-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20181116 06:35:29< irker735> wesnoth/wesnoth:master mattsc b1ad86aeba Unit tests: change attack_depth to leade AppVeyor: 1/4 builds failed 20181116 06:35:30< irker735> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/builds/20341564 20181116 08:37:01<+wesdiscordbot> @jyrkive By the way, were you planning to work on OpenGL at all anymore for 1.15, or is that not happening anymore? 20181116 08:38:10<+wesdiscordbot> Yes, I'm planning to work on it. (But take the "for 1.15" out. I'm not exactly targeting any particular release.) 20181116 08:38:42<+wesdiscordbot> There has just been higher-priority stuff, such as forward-porting the 1.14 commits, and now implementing a better way to position the battery widget. 20181116 09:08:57<+wesdiscordbot> But why 20181116 09:09:41<+wesdiscordbot> Your question is too unclear. What exactly are you asking? 20181116 09:09:44<+wesdiscordbot> Why would you proceed with the OGL work, I mean. 20181116 09:09:55<+wesdiscordbot> Why wouldn't I? 20181116 09:10:14<+wesdiscordbot> Open source development is all about doing things I feel like doing. 20181116 09:11:59<+wesdiscordbot> That's true. But if series 2 works out, won't your work then just be replaced :/ 20181116 09:12:45<+wesdiscordbot> Sure. But I don't care. 20181116 09:12:50<+wesdiscordbot> Alright. 20181116 09:13:47<+wesdiscordbot> I just don't want you to feel like your work was a waste later. šŸ˜¦ 20181116 09:14:12<+wesdiscordbot> (I'm not saying that to be passive-aggressive, FTR) 20181116 09:14:51< irker735> wesnoth/wesnoth:master gfgtdf 1799cea6de don't send whiteboard data to older clie AppVeyor: All builds passed 20181116 09:18:57<+wesdiscordbot> @Vultraz FWIW, I wouldn't hold back on any development towards Wesnoth 1.x Even if haldric will be successful, I think it is very likely that it won't replace Wesnoth 1.x series at all. There will always be devs who'd rather develop for 1.x, and there will always be players who want to play 1.x over whatever haldric will turn out to be. Thus, I really think haldric should not be the reasoning for not implementing new 20181116 09:18:57<+wesdiscordbot> features to wesnoth 1.x. 20181116 09:19:20<+wesdiscordbot> I have also considered this. 20181116 09:19:32<+wesdiscordbot> especially given the art discussion yesterday 20181116 09:19:53<+wesdiscordbot> Also true. 20181116 09:20:07<+wesdiscordbot> That would mean splitting the team permanently, though. 20181116 09:20:14<+wesdiscordbot> Probably not a bad idea, I suppose 20181116 09:21:31<+wesdiscordbot> I found myself feeling my work being wasted at a previous workplace. I had developed a mobile game there essentially alone; however, later on they hired another programmer as a CTO (he was above me in the hierarchy). He strongly disagreed with a software design decision of mine, and rewrote the whole game from scratch without asking my permission. I decided to give in and work on his codebase from then on instead of 20181116 09:21:32<+wesdiscordbot> mine. The decision in question was how to implement validation of player moves in the server. My approach was to simply not validate at all initially, and if validation is later wanted, then duplicate the entire game logic in the server. He wanted to instead share the game logic between client and server (both were written in C#, so it was possible). To top it all off, afterwards it turned out that I had been right. The resulting 20181116 09:21:32<+wesdiscordbot> "shared game logic" was extremely fragile in the client and very difficult to use. It was a complete nightmare to add more cards to the game while trying not to introduce bugs in corner cases. And this is not even the primary reason why I quit from there. I'm glad to have left them behind... 20181116 09:23:21<+wesdiscordbot> In the case of Haldric, I'm at least perfectly aware that the code might never be used while I'm writing it, instead of having to deal with someone else arriving afterwards and actively throwing it away. 20181116 09:24:09<+wesdiscordbot> Will you continue to work off the foundation I set up? 20181116 09:24:27<+wesdiscordbot> No. I plan to rebase the OpenGL branch on the current master. 20181116 09:24:42<+wesdiscordbot> I see 20181116 09:25:08<+wesdiscordbot> I can't deny I myself feel like my work was wasted. šŸ˜¦ 20181116 09:25:35<+wesdiscordbot> Yeah, I realize that. I know, from experience, that it's not a good feeling. šŸ˜¦ 20181116 09:29:13<+wesdiscordbot> I like to believe that no work is wasted, be it only for the things you have learned during the process. (That of course gets a bit different if you already know everything there is to know about the matter, but since I am a beginner in pretty much everything, it always applies for me :P) 20181116 09:29:47<+wesdiscordbot> I definitely learned a lot 20181116 09:30:07<+wesdiscordbot> I figure what I learned in my case is: never, ever let your game state run ahead of rendering. 20181116 09:33:21-!- valdar [~atarocch@93.56.172.28] has quit [Ping timeout: 252 seconds] 20181116 10:05:25<+wesdiscordbot> FTR, a major thing we've been discussing for Haldric is the upgrade of many art assets. That's where the Patreon discussion comes in, and also why I'm considering it is perhaps a more likely possibility than before that series 1 will end up remaining as its own thing as opposed to being completely replaced. 20181116 10:08:52<+wesdiscordbot> LordBob has already produced some HD grass that looks goddamn amazing. 20181116 10:10:07-!- valdar [~atarocch@93.56.172.28] has joined #wesnoth-dev 20181116 10:27:16<+wesdiscordbot> Yeah it's seems it will become much more that just a straight port. 20181116 11:59:57-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20181116 12:18:00-!- irker735 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181116 12:18:07-!- irker884 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181116 12:18:07< irker884> wesnoth/wesnoth:master Celtic Minstrel 2f9bca43af Fix erroneous deprecation AppVeyor: All builds passed 20181116 12:30:45<+wesdiscordbot> The idea I'm currently tossing around is around Christmas we can release 1.15.0 alongside a Patreon launch and some more info (and one or two screenshots) about Series 2. 20181116 12:32:02<+wesdiscordbot> 1.15.0 wouldn't have much to offer at all... 20181116 12:32:50<+wesdiscordbot> I know. 20181116 12:33:13<+wesdiscordbot> But you guys seem to want a release. 20181116 12:33:26<+wesdiscordbot> I don't, for the record. 20181116 12:33:33<+wesdiscordbot> Hm 20181116 13:33:32< celmin|sleep> I want a release, yes. 20181116 13:33:40< celmin|sleep> But it doesn't have to be right now. 20181116 13:36:12-!- celmin|sleep is now known as celmin|away 20181116 13:36:45< celmin|away> It would be good to get some actual new content in, such as sigurd's drake campaign or my random cave map or both. 20181116 13:36:57< celmin|away> Is there any other new content pending? I don't know of any... 20181116 13:37:03< celmin|away> Not sure if the new units count. 20181116 13:37:46< celmin|away> I mean, they're available in the map editor, so they might? But probably most players won't be using the map editor? 20181116 14:07:59<+wesdiscordbot> there's the schema validation, it matters to UMC authors 20181116 14:30:17<+wesdiscordbot> I hope he'll write a WML test this time 20181116 14:30:34<+wesdiscordbot> I asked him before but he said something about not being able to 20181116 14:39:45<+wesdiscordbot> It's not pnlay about spliting the deveoplment teams, it's also about splitting the player base. 20181116 14:40:19<+wesdiscordbot> I saw yesterday only 4 open games when I joined the MP server. Admittedly, that's the lowest I ever saw. 20181116 14:40:32-!- TC01 [~quassel@venus.arosser.com] has quit [Ping timeout: 244 seconds] 20181116 14:40:50<+wesdiscordbot> I would guess ~100 players can be expected when joining in average 20181116 14:42:21<+wesdiscordbot> I've seen 100+ players and 30+ games more than once when I've been online on the weekend 20181116 14:48:54-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Read error: Connection reset by peer] 20181116 14:50:26-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev 20181116 14:55:17<+wesdiscordbot> Me too. My point is that the playerbase is also with that still pretty small 20181116 14:55:29<+wesdiscordbot> Vultraz said sth like old players go, ne wcome 20181116 14:55:57<+wesdiscordbot> *new come, but we just had the big 1.14 releaseā€¦ so I don't expect a too big grow 20181116 14:57:26<+wesdiscordbot> I understood the original idea as to have both series 1 and 2 coexist for a while but to switch over to 2 in the long run 20181116 14:58:16<+wesdiscordbot> I don't think this is something we can control. This is a desicion that each and every one of both players and devs will make for their own. 20181116 14:58:20<+wesdiscordbot> the original idea was to have 2.0 be a straight port to a new engine 20181116 14:58:31<+wesdiscordbot> *was also 20181116 14:58:45<+wesdiscordbot> I think no one is in control about this. We can't really tell people that they're not allowed to work on something. It would be strongly against the principles of open source. 20181116 15:01:07<+wesdiscordbot> My original proposal was to treat series 2 was a development series until it was done, then once actually released as a new stable series, series 1 would undergo the usual treatment that an obsolete stable series has 20181116 15:01:23<+wesdiscordbot> More specifically, as a longer development series like Wesnoth 0.x was 20181116 15:02:04<+wesdiscordbot> Because dropping something fresh out of the oven in production and replacing the old thing immediately is... just a terrible idea in general 20181116 15:02:22<+wesdiscordbot> Hence Wesnoth goes through alpha, beta and release candidate phases before GA 20181116 15:03:34<+wesdiscordbot> There are some differences which make some developers not want to work on the 2.x rewrite (such as that I'm here mostly because I want to have some C++ experience to show in GitHub, and thus working on 2.x would be pointless for me). 20181116 15:06:46<+wesdiscordbot> Incidentally the issue with players not caring about development releases does factor into this. People just don't get excited at development releases because no matter what you do, it still looks and feels exactly like the previous stable series short of a major UI or content overhaul (and there hasn't been a content overhaul ever) 20181116 15:07:34<+wesdiscordbot> People work on that they want, true. But the idea of split development teams which I saw mentioned, already at this state of thingsā€¦ Honeslty, there has been a lot of different opinions on wensoth 2, and I see it more as an hidden conflict becomming biggerā€¦ we shouldn't fall due to that. I think we do have some influrence on wether 1.x will remain permanently ā€¦ nobody get's the idea to continue on 1.12.x 20181116 15:07:52<+wesdiscordbot> So unless you're a content developer, development releases just don't have anything new to offer and very few people who play the full game (meaning: campaigns AND multiplayer) are generous/wealthy enough to donate their time to finding and reporting bugs that are sometimes extremely frustrating in nature 20181116 15:08:51<+wesdiscordbot> Wesnoth 2 happens to be a great opportunity to actually break out of that tradition since no-one is doing much to change Wesnoth 1 20181116 15:09:33<+wesdiscordbot> Or if they do they do it at an extremely slow pace that takes multiple stable series to materialize (see also: UtBS, Dunefolk) 20181116 15:09:51<+wesdiscordbot> Because I feel like too many people here believe they're shipping software piecewise 20181116 15:09:58<+wesdiscordbot> Rather than shipping a game 20181116 15:10:38< Soliton> how would wesnoth 2 break out of that tradition? 20181116 15:10:44<+wesdiscordbot> (Bear in mind that when I say "multiple stable series", since 1.9.x that means at least 2-3 years in development) 20181116 15:11:03<+wesdiscordbot> New people in the development team with a different mindset basically 20181116 15:11:32<+wesdiscordbot> I think the idea of working on higher resolution graphics (which the mainline engine doesn't even support right now) was mentioned above 20181116 15:11:44<+wesdiscordbot> They are also doing a new HUD, if I remember correctly 20181116 15:11:56< Soliton> maybe i didn't understand what you're referring to with tradition. i thought you mean getting people interested in dev versions. 20181116 15:12:14<+wesdiscordbot> I mean both things, one thing leads to the other 20181116 15:12:33<+wesdiscordbot> I just explained the player base doesn't care about development versions because they are buggy and don't have anything new for players to bring to the table 20181116 15:12:51<+wesdiscordbot> Except for those few exceptions that again took forever to be completed 20181116 15:12:58<+wesdiscordbot> And one of them was even removed once 20181116 15:13:01<+wesdiscordbot> Personally, I doubt that even significant art improvements or UI changes would drive people to try unstable versions. 20181116 15:13:43<+wesdiscordbot> It's not just cosmetic changes though 20181116 15:14:36<+wesdiscordbot> There's still that issue where Wesnoth is extremely unfriendly to high resolution screens in terms of both visuals (without zooming into the map and increasing font size, which causes a huge number of issues because of the engine's design) and performance (CPU rendering) 20181116 15:15:44<+wesdiscordbot> For all intents and purposes the AI engine and the WML/Lua engine (and the latter is quite debatable) are the only things in mainline that were done right and don't have inherent flaws of some sort hindering both development and UX 20181116 15:16:08<+wesdiscordbot> We've been working on fixing the UI for ten years for reference 20181116 15:16:48<+wesdiscordbot> Not because we don't know how to come up with a usable UI, but because we don't know how to write a UI framework that's both maintainable and performant 20181116 15:17:05<+wesdiscordbot> Performant in the context of the current CPU-based engine 20181116 15:17:36<+wesdiscordbot> Now bear in mind the average player of a game doesn't stick around for longer than a year 20181116 15:17:38<+wesdiscordbot> More like "because when someone (= mordante) was willing to do the work to switch to a different GUI framework, he decided to write his own instead of using an existing one" 20181116 15:18:27<+wesdiscordbot> I might also add that from the 7 people that have been contributing to project haldric so far, only 3 have been working on wesnoth 1 before. if I'm not mistaken. The majority of what has been done does not come from previous wesnoth devs. And yes, as shadown mentioned, were completely reworking UI, and also address the low resolution of current assets. Not to mention that we can fully make use of GPUs with Godot. IIRC 20181116 15:18:28<+wesdiscordbot> Wesnoth 1. is entirely, or at least almost entirely done by CPU. 20181116 15:18:43<+wesdiscordbot> (Also, to clarify, the rendering code was done right... for 2005. We've long since abandoned 1024x768 screens and systems without a capable built-in GPU of some sort.) 20181116 15:19:05<+wesdiscordbot> (I disagree. IMHO; OpenGL should have been used even back in 2005.) 20181116 15:19:40<+wesdiscordbot> Well, I think the OpenGL argument until around 2011 came down to four letters 20181116 15:19:43<+wesdiscordbot> Mesa 20181116 15:20:13< zookeeper> i think it'd be reasonable to specifically state/agree that development of either project should not be intentionally impeded at any point. as in, even if haldric became super awesomely successful, then people who want to switch to haldric shouldn't get to decide on whether wesnoth development stops if there are still people who wish to keep working on it. i mean, to make it clear that 20181116 15:20:13< zookeeper> even if most wesnoth devs/admins eventually switched to haldric, they, as an extreme example, wouldn't actually lock down the wesnoth repo or anything silly like that. 20181116 15:20:46<+wesdiscordbot> My laptop with a Radeon HD 3200 was basically crippled back in the day (2008-2010) because Mesa and the rest of the graphics stack didn't fully support its features, and the proprietary drivers were crashy af 20181116 15:21:09<+wesdiscordbot> Back in the day most Wesnoth devs were using Linux 20181116 15:21:40-!- irker884 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20181116 15:21:52<+wesdiscordbot> 2D rendering with OpenGL is rather simple, though. Just two triangles per sprite, one shader, stuff like that. Any driver that's not hopelessly broken should be able to support that much. 20181116 15:22:04<+wesdiscordbot> There were like 2-3 developers at a given time using Windows or macOS, then the trend reversed when the team shrank down around 1.12 20181116 15:22:28<+wesdiscordbot> Why did the team shrank down? 20181116 15:22:33<+wesdiscordbot> Because people left :p 20181116 15:22:56<+wesdiscordbot> remember this post here: https://forums.wesnoth.org/viewtopic.php?t=42912 20181116 15:23:58<+wesdiscordbot> Also jyrkive you say that but back in the day even simple things worked unreliably on Linux with Mesa 20181116 15:24:43<+wesdiscordbot> All right. I don't know much about those days. I have only been using GNU/Linux as my secondary operating system, starting from 2009. 20181116 15:24:45<+wesdiscordbot> NVIDIA was basically the only source of fully stable drivers... which a vocal segment of our audience at the time wouldn't even touch since they're proprietary 20181116 15:25:11<+wesdiscordbot> There was a massive boost in development for Mesa when certain companies took an interest on Linux 20181116 15:25:55<+wesdiscordbot> I think these days most recent AMD graphics card work with it out of the box 20181116 15:26:22<+wesdiscordbot> In 2009, support for mine (which wasn't even particularly new) was still a work in progress 20181116 15:26:43-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Ping timeout: 245 seconds] 20181116 15:33:04-!- irker196 [~irker@wesnoth/umc-dev/bot/shikadibot] has joined #wesnoth-dev 20181116 15:33:04< irker196> wesnoth/wesnoth:master newfrenchy83 c5df8edd17 Update udisplay.cpp AppVeyor: vs2015/Release Failed 20181116 15:33:04< irker196> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20355585 20181116 15:45:38<+wesdiscordbot> Well, ... I think the only reason why 1.x would permanently be used after (and in case if) the 2nd series becomes successfulā€¦ if there are enought devs and players which think wesnoth 2 is worse. We are now at the stage that we can take care that this isn't the case. I think series 2 has a good starting point to replace 1.x because players will like a new UI, as for the rest ā€¦ Add-ons not being able to be ported in any 20181116 15:45:39<+wesdiscordbot> way is a big point. Guess it's the turning point. 20181116 15:47:51<+wesdiscordbot> If you're wondering what we're currently experimenting with in terms of UI. There are also some other ideas in the pipeline, but that's the current look. (With old terrains though) 20181116 15:47:51<+wesdiscordbot> https://cdn.discordapp.com/attachments/259976436490829825/513017357346799626/Screenshot-2018-11-16-16-42.png 20181116 15:48:03<+wesdiscordbot> the current direction of 2.0 is to create an completely new UMC API, and also replace most/all of the current art with higher res art. if it works, it would be great, but it's a ton of work both in terms of getting mainline done as well as what it would take to port existing add-ons(which is why the patreon was brought up now). 20181116 15:50:18<+wesdiscordbot> @sevu ^this of course could be one reason to prefer Wesnoth 1.x series over Wesnoth 2, not because it's worse, but because there are certain UMC campaigns that one want's to play. Wesnoth 2.0 series will most likely have different content, so there is no reason why not to play both. 20181116 15:51:51<+wesdiscordbot> Credit to bitron for building off my concept design for a new UI and improving it, ftr 20181116 15:51:51<+wesdiscordbot> AE for example would be unlikely to ever get ported, I'd expect 20181116 15:52:16<+wesdiscordbot> Ageless Era is broken by design šŸ˜ƒ 20181116 15:52:20<+wesdiscordbot> I would hope not. If only because weā€™d design around it so it wasnā€™t needed 20181116 15:52:21<+wesdiscordbot> That's a dragon.. 20181116 15:53:01<+wesdiscordbot> Or to put it in better terms, Ageless Era wouldn't exist if Wesnoth wasn't broken by design 20181116 15:53:02<+wesdiscordbot> Indeed 20181116 15:53:16<+wesdiscordbot> AE has a ton of coding for unit ability and a ton of art for the multitude of different factions. it'll be a ton of work regardless of how much better the UMC API is. 20181116 15:53:32<+wesdiscordbot> AE is a compilation 20181116 15:54:03<+wesdiscordbot> I thought Arts would be the thing which could trivally be taken overā€¦ For most of mainline, losing the one or other campaing whouldn't be a big loss, as the basically have all the same simple gameplay (Can be added latter like many games add DLCs over time). I'm more worrying about the big things like RPG worls from UMC. (AE is a hell which only exists because it doesn't need any work, it's a collection of all worse code 20181116 15:54:04<+wesdiscordbot> whose Eras died. It has also been brought up to solve the problem which AE somes different in a future version) 20181116 15:54:59<+wesdiscordbot> Wouldn't it make total sense to be able to allow multiple eras for a multiplayer match to choose factions from? This would eliminate the need for such bundles as AE is one. 20181116 15:55:28<+wesdiscordbot> the big umcs are AE, EoMa, LotI, some of kwandulin's stuff, XP mod 20181116 15:56:20<+wesdiscordbot> if multiple era selection is allowed, eras themselves don't really seem to make sense 20181116 15:56:27<+wesdiscordbot> I'm not sure how inferno8 feels about porting his stuff 20181116 15:56:31<+wesdiscordbot> just have a checkbox for what factions you want for the match 20181116 15:56:31<+wesdiscordbot> Oi. AtS and IftU too 20181116 15:56:40<+wesdiscordbot> I'm going off DL count 20181116 15:56:47<+wesdiscordbot> cri 20181116 15:56:51<+wesdiscordbot> additionally 20181116 15:57:16<+wesdiscordbot> We can work with inferno 8 to help him port his stuff if necessary 20181116 15:57:22<+wesdiscordbot> Same with shadowm. 20181116 15:57:24<+wesdiscordbot> we have shadowm here right now, she usually speaks up for herself for her own projects 20181116 15:57:27<+wesdiscordbot> Not dugi. 20181116 15:57:33<+wesdiscordbot> agreed, we both would be willing to help shadowm 20181116 15:57:47<+wesdiscordbot> Indeed 20181116 15:58:21<+wesdiscordbot> btw as far as new content for a 1.15 release 20181116 15:58:35<+wesdiscordbot> my TSG rewrite could sort of be one (?) 20181116 15:58:47<+wesdiscordbot> it's not reaaalllllly a rewrite, but at least is something 20181116 15:58:54<+wesdiscordbot> although I think the target is 1.14.6 right now 20181116 15:59:13<+wesdiscordbot> perhaps 20181116 16:00:01<+wesdiscordbot> the dunefolk liminal change would also be something players would be able to at least see. 20181116 16:00:44<+wesdiscordbot> I wonder if sth. like strange legacy could be ported 20181116 16:01:37<+wesdiscordbot> If we start thinking only in terms of porting, we'll severely limit ourselves before we even begin. 20181116 16:02:55<+wesdiscordbot> Agreed. Though completely ignoring that there is an UMe have is the other extremum 20181116 16:02:56<+wesdiscordbot> Agreed. I'd like Haldric to be a new platform for new content, not something that's build to accept old content. 20181116 16:03:21<+wesdiscordbot> at this point, porting really only consists of the new API's functionality anyway. There's really not much that can be carried over from 1.x to 2.0 besides music. 20181116 16:03:44<+wesdiscordbot> and things like unit stats, ability configs, etc 20181116 16:06:53<+wesdiscordbot> units are tivial. abilities/specials either as well or they work around over tricks in the current implementation which doesn't work but that's fine. 20181116 16:07:41<+wesdiscordbot> I wonder if the logic written in events could somehow be ported 20181116 16:08:30<+wesdiscordbot> tbh probably the most difficult thing (for someone like me) would be to recreate art assets for the 2.0 version 20181116 16:08:37<+wesdiscordbot> since the resolution would be different 20181116 16:08:42<+wesdiscordbot> I might not even be capable of it 20181116 16:09:20<+wesdiscordbot> code is annoying, but if the new API really is better, it should be relatively quick 20181116 16:14:42<+wesdiscordbot> Somewhat, theoretically. I guess it's a matter if that person maintaining it has a few weeks to rewrite the stuff. I rewrote the Era of Myths files a few years ago to the new animation syntax ā€¦ took several weeks. And it's basically trivial. 20181116 16:20:30<+wesdiscordbot> I don't know if I can ask in this chat. But whats UMC stands for? Seems like you guys are planning a major update? 20181116 16:20:40<+wesdiscordbot> User Made Content 20181116 16:21:03<+wesdiscordbot> Tks @Pentarctagon 20181116 16:23:44<+wesdiscordbot> Well, what shadowm brought up earlier about Lau/WML being one of the points which have been done right in wesnoth, assuming for a moment that the code could be reused in series 2, that would resolve this. I don't let the argument count that from a programmers mindset some things are looking awkward with it as a languageā€¦because it'd doing it's job fine so far. There's no rason why Yaml must replace WML besides that godot 20181116 16:23:44<+wesdiscordbot> offers something for Yaml. 20181116 16:27:19<+wesdiscordbot> it would be good to implement the new API in a way that could support full rolling releases, which I've been told is basically impossible with 1.x. 20181116 16:28:56-!- TC01 [~quassel@venus.arosser.com] has joined #wesnoth-dev 20181116 16:32:42-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev 20181116 16:33:33<+wesdiscordbot> what would be the requirements for such a system? šŸ¤” 20181116 16:37:20<+wesdiscordbot> Sorry, sevu, we are not keeping WML 20181116 16:38:51<+wesdiscordbot> reason being, if I understood that correctly, that WML wasn't designed to be a scripting language, thus has some flaws that we'd like to avoid having in series 2 as well. 20181116 16:39:34<+wesdiscordbot> That is true regarding scripting, yes. 20181116 16:39:56<+wesdiscordbot> Itā€™s also just rather awkward as a data language even and a standard format is better 20181116 16:43:14< Ravana_> josteph: first IRC connection failed, then USB device too, and then systemd was taking away all RAM. Not going to debug it further currently 20181116 16:43:34<+wesdiscordbot> @Bitron For the yaml I'm not sure, but for the lua the requirement would be that it should be possible to execute different logic based on the API level the user or UMC has set. 20181116 16:46:28<+wesdiscordbot> I think we all gread that things like [foreach] won't be a loss. But "just rather awkward" as a data description language is not an argument, I think that's one of the reason why you have such a huge disagreement with celmin if you argue with such general things. I'm also not sure why you think so, it's basically key=vale, encapsulated by tags. getting rid of [+tags] (note the +) would probably no loss. "a standard format" 20181116 16:46:28<+wesdiscordbot> would have been a good argument if there wouldn't already have been such a huge amount of code written in WML yet 20181116 16:46:47<+wesdiscordbot> having that would allow making breaking changes without breaking compatibility with exist UMC until that API level is removed entirely 20181116 16:49:09<+wesdiscordbot> that in turn would allow people to use the latest release without losing access to their current UMC/saves/etc 20181116 16:53:19<+wesdiscordbot> For balance changes we didn't found a solution 20181116 16:55:47<+wesdiscordbot> But there was this guy in IRC from some other game development team (I don't remember the name, but he was looking ho the MP stuff was done here) who told that they solved the problem in the way that the application which distributers ship / users install downloads then all the stuff ā€¦ i.e. wesnoth would not deliver a core, but threat it like add-ons. 20181116 16:55:50< Ravana_> by designing units to be sent on demand 20181116 16:58:49<+wesdiscordbot> the simple solution would be to just have like: data/1/units/... data/2/units/... but there's probably a better way to do that. 20181116 17:07:28<+wesdiscordbot> Nice idea. If working with lua composing the final unit based on diferent files should be simple. To overload methods and so. But I guess that if only logic will change and not all the assets, unit should come first in the path? data/units/1/... 20181116 17:08:09<+wesdiscordbot> depends on how it's organized, really 20181116 17:08:23<+wesdiscordbot> could just have data/images and such separately. 20181116 17:12:16< Soliton> you can put balancing changes on the add-on server already. 20181116 17:13:03<+wesdiscordbot> Whats the plan in changing the UMC @Pentarctagon, there's any post explaning this that I can read to know more about? 20181116 17:13:40< Soliton> unit definitions are purposely not dynamic otherwise cheating that currently simply causes OOS would be very difficult to prevent/detect. 20181116 17:15:44<+wesdiscordbot> @sevu: FYI: https://github.com/wesnoth/wesnoth/commit/073f70b 20181116 17:16:00<+wesdiscordbot> That would take care of your problem with using a deifferent default AI in a core. 20181116 17:16:19<+wesdiscordbot> And I think it could be cherry-picked to 1.14 without causing trouble. 20181116 17:16:24<+wesdiscordbot> @TheRatazana Right now the direction of 2.0 is there will be no compatibility with 1.x's UMC API. 20181116 17:17:29<+wesdiscordbot> Ok 20181116 17:45:56-!- gfgtdf [~androirc@tmo-104-123.customers.d1-online.com] has joined #wesnoth-dev 20181116 17:46:56< gfgtdf> celmin: what did you want to know about the mapgen? 20181116 17:48:49< gfgtdf> Celmin: that commit does contain an example lua code which is (except that that is squashed all to one file) basically exactly what i want o do in my add-on. 20181116 17:49:41< gfgtdf> so if you want to change it make aure that example code still works, and with equal performance. 20181116 17:55:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20181116 17:59:27< Ravana_> ageless started out as multiera, but now it has more important function of keeping abandoned eras available 20181116 18:03:24< gfgtdf> iirc ageless also balances them against each other. 20181116 18:03:44< irker196> wesnoth/wesnoth:master mattsc 81f3d2fcd2 AI: move default algorithm definition fr AppVeyor: vs2015/Release Failed 20181116 18:03:45< irker196> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20359868 20181116 18:04:01< Ravana_> that is more recent tend, only started around 2016 20181116 18:04:03-!- gfgtdf [~androirc@tmo-104-123.customers.d1-online.com] has quit [Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )] 20181116 18:36:11<+wesdiscordbot> @sevu If 2.0 doesn't use WML the question is whether it provides an upgrade path... if it uses say XML and there's an automatic converter, that can work 20181116 18:42:16<+wesdiscordbot> @josteph it uses yaml. It's probably possible to write a converter for that too, but mind that some things will be stored differently, like abilities and such. 20181116 18:48:57<+wesdiscordbot> 2.0 also doesn't use macros, so that would also be something to account for 20181116 18:50:10< Ravana_> macros as wml has them is certainly something to get rid of 20181116 19:05:51< zookeeper> _maybe_ worth noting is that if you could provide a converter for unprocessed WML -> yaml then people could just run their code through the wesnoth preprocessor and then convert _that_. of course the code would be a mess, but it should still work. assuming the APIs were identical enough that a converter like that could be written, anyway; i don't know what the plan is WRT that. 20181116 19:07:00<+wesdiscordbot> I think that at most unit definitions and the like might be convertible. But not code. 20181116 19:07:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20181116 19:15:11<+wesdiscordbot> lua code would be all that could possibly work, and so far I haven't heard that there's any intention to design 2.0's lua API in a way that would be compatible. 20181116 20:32:57< irker196> wesnoth/wesnoth:master newfrenchy83 c5df8edd17 Update udisplay.cpp AppVeyor: 1/4 builds failed 20181116 20:32:58< irker196> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20355585 20181116 20:58:04-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 250 seconds] 20181116 21:13:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20181116 22:11:34-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20181116 23:03:44< irker196> wesnoth/wesnoth:master mattsc 81f3d2fcd2 AI: move default algorithm definition fr AppVeyor: 2/4 builds failed 20181116 23:03:45< irker196> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/builds/20359868 20181116 23:03:46< irker196> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/builds/20359869 20181116 23:30:56-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20181116 23:57:02-!- sevu [~sevu@p5485415E.dip0.t-ipconnect.de] has joined #wesnoth-dev --- Log closed Sat Nov 17 00:00:37 2018