--- Log opened Sat Apr 07 00:00:34 2018 20180407 00:01:55< mattsc> celticminstrel: Hi. Do you have a couple minutes for me to ask some questions on the recall/recruit bug? 20180407 00:02:19< mattsc> If so, let’s start here: https://github.com/wesnoth/wesnoth/blob/master/src/ai/default/recruitment.cpp#L257 20180407 00:04:55< irker393> wesnoth/wesnoth:1.14 loonycyborg 32178aab81 flatpak-builder manifest: cleanup a buil AppVeyor: All builds passed 20180407 00:08:46< celticminstrel> I actually had a question or two for you too. 20180407 00:09:00< celticminstrel> For example, is there a benefit to using id instead of [filter] in MicroAIs? 20180407 00:09:16< mattsc> No. That’s entirely for backward compatibility. 20180407 00:09:20< celticminstrel> 'kay 20180407 00:10:01< celticminstrel> So this is the execution function of the recruitment CA. 20180407 00:10:10< mattsc> Right. 20180407 00:10:25< mattsc> So, recalls get added to the “recruit” list here. 20180407 00:10:44< mattsc> But they get added by type id. not as individual units. 20180407 00:11:00< mattsc> So then the function does it’s evaluation, based on the instructions it has etc. 20180407 00:11:08< mattsc> Next: line 380 20180407 00:11:42< mattsc> At this point it has chosen the recuit it wants to make, and in 381, it checks if there’s a unit to recall of that type. 20180407 00:12:12< mattsc> That’s done in get_appropriate_recall() in l.477 etc. 20180407 00:12:33< mattsc> The important bit is l.513 20180407 00:12:48< mattsc> Sorry. l.510. 20180407 00:13:04< celticminstrel> Hmm. 20180407 00:13:09< mattsc> It only adds units if the recall value is less than the recall cost 20180407 00:13:22< mattsc> Recall value for a unit without XP is essentiall its cost. 20180407 00:13:47< mattsc> So, now if we jump back to after l.381 ... 20180407 00:14:10< mattsc> It can happen that no recall was considered worthy, and if that unit type is not on the recruit list, then the CA is blacklisted. 20180407 00:14:25< mattsc> So now for my question: 20180407 00:14:48< mattsc> The obvious thing to do is to check whether a unit is worth recalling before adding its type to the recruit list. 20180407 00:15:15< mattsc> But if we do that, because of the way it is set up, the is_it_worthy evaluation needs to be done twice. 20180407 00:15:55< mattsc> So, do we do it twice (because it does not take much time) or do we change how this is set up (in which case I likely need help)? 20180407 00:16:27< celticminstrel> The thing twice is get_appropriate_recall()? 20180407 00:16:37< mattsc> Pretty much, yes. 20180407 00:16:57< celticminstrel> Hmm. 20180407 00:17:06< mattsc> Exactly. 20180407 00:18:21< celticminstrel> Isn't this at least partly a problem with the evaluation function not even considering the possibility of a recall? 20180407 00:18:22< mattsc> Also to keep in mind is that this is only half of the problem (this is the one to be fixed for 1.14). 20180407 00:18:40< mattsc> Right, that is the second problem. 20180407 00:18:44< celticminstrel> Ah. 20180407 00:18:55< mattsc> However, we need to fix one in 1.14, and both in master. 20180407 00:19:07< celticminstrel> Why is that? 20180407 00:19:22< mattsc> Because of the feature freeze. 20180407 00:19:49< celticminstrel> Obviously, but what's the reasoning behind whether the two changes break the freeze... 20180407 00:20:37< mattsc> Problem 1: If the AI decides that it wants to recruit a unit of the recall list and then decides it’s not worthy, all recruiting is stopped. 20180407 00:20:43< mattsc> That’s a bug and needs to be fixed. 20180407 00:20:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180407 00:21:12< mattsc> Problem 2: if the AI cannot do anything but recall “unworthy” units, it won’t recall those. 20180407 00:21:39< mattsc> That’s a feature, and I can come up with all kinds of scenarios how UMC authors might be using that without even knowing it. 20180407 00:22:00< mattsc> It’s not a desirable feature as default, but a feature nevertheless. 20180407 00:22:35< celticminstrel> Okay, so the current sequence of events is... 1. Get list of possible recruit types, accounting for recruit, extra_recruit, and recall list. 2. Decide which available type is the best. 3. Check if that type is on the recall list. 4... 20180407 00:22:45< mattsc> In fact, what I would suggest to do about this in master is to add a configuration parameter with which you can switch between the two. 20180407 00:22:53< celticminstrel> The 4... is only there to indicate that there's more I've omitted. 20180407 00:23:01< celticminstrel> Switch between the two what? 20180407 00:23:25< mattsc> Do not recall peasant for 20 gold if you cannot do anything, or do it anyway of that is the only thing you have left to do. 20180407 00:23:30< celticminstrel> By configuration parameter, do you mean new keys in [recruit] or a new aspect? 20180407 00:23:31< mattsc> *peasants 20180407 00:23:37< mattsc> new aspect 20180407 00:23:53< mattsc> celticminstrel: I have to apologize, but I am being called away. 20180407 00:24:01< mattsc> I’ll be back … don’t know, in an hour ot so. 20180407 00:24:44< celticminstrel> Hopefully I'll be around too when you get back. 20180407 00:37:14< irker393> wesnoth: stevecotton wesnoth:1.14 36a78d885606 / data/campaigns/Legend_of_Wesmere/lua/wml_tags.lua: LoW S09 Lua cleanup: fix a line that was missed in 3783693cb (#2835) https://github.com/wesnoth/wesnoth/commit/36a78d8856069b5fcfcab7f736f4708be23b30fb 20180407 00:38:54-!- gfgtdf [~chatzilla@x4e363967.dyn.telefonica.de] has joined #wesnoth-dev 20180407 00:39:10< mattsc> celticminstrel: If not, we can pick it up some other tiime/day. Btw, Step 2. in your list can be as simple as a random choice. Which means the bug may or may not appear even starting from the same save. 20180407 00:39:31< irker393> wesnoth: Celtic Minstrel wesnoth:fix_2833 884f88cb89b4 / data/campaigns/Legend_of_Wesmere/scenarios/chapter3/09_Bounty_Hunters.cfg: fixup incorrect function https://github.com/wesnoth/wesnoth/commit/884f88cb89b4e4bb0eeec59e735c20869caa58a8 20180407 00:46:22-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20180407 01:13:11<+discordbot> celmin: please commit the LoW fix to master too 20180407 01:24:28<+discordbot> nevermind, I did it myself 20180407 01:24:32< irker393> wesnoth: stevecotton wesnoth:master 2b406a37ec00 / data/campaigns/Legend_of_Wesmere/lua/wml_tags.lua: LoW S09 Lua cleanup: fix a line that was missed in 3783693cb (#2835) https://github.com/wesnoth/wesnoth/commit/2b406a37ec009e4a13251035267356b4d82ab1f1 20180407 01:41:00< mattsc> celticminstrel: I’m back, at least for a while. 20180407 01:58:12-!- gfgtdf_ [~chatzilla@x4e36316a.dyn.telefonica.de] has joined #wesnoth-dev 20180407 02:00:13-!- gfgtdf [~chatzilla@x4e363967.dyn.telefonica.de] has quit [Ping timeout: 260 seconds] 20180407 02:00:21-!- gfgtdf_ is now known as gfgtdf 20180407 02:00:51-!- timotei__ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 246 seconds] 20180407 02:09:19< celticminstrel> mattsc: I'm back too now. 20180407 02:09:36< mattsc> celticminstrel: hi 20180407 02:11:11< mattsc> So do you have an opinion or any advice on how to go ahead with this. 20180407 02:12:14< mattsc> Is just calling the value check in get_appropriate_recall() twice acceptable for something like this? 20180407 02:12:33< celticminstrel> Hmm... it's iterating over the whole recall list... 20180407 02:12:51< mattsc> Well, yes, you wouldn’t do that for each recall. 20180407 02:13:56< mattsc> But you do have to iterate over it twice. So it’s 2*n rather than n^2. 20180407 02:14:44< mattsc> Obviously it would be better to avoid that, but I don’t see how to do this without changing how this evaluation is organized. 20180407 02:14:50< celticminstrel> Feels like the logical thing to do here amounts to "filter twice"... 20180407 02:15:05< celticminstrel> So iterate over the recall list, copy points to "appropriate" recalls of any type into a new list... 20180407 02:15:27< celticminstrel> Then where it currently calls get_appropriate_recall, instead go through that list for a unit of the chosen type. 20180407 02:15:37< mattsc> Agreed. 20180407 02:15:58< mattsc> I’m a bit unsure on how to code that though. 20180407 02:16:13< mattsc> In C++, I mean. In pseudocode that’s no problem. 20180407 02:16:16< celticminstrel> Yeah, it'd be quite a significant change to the code. 20180407 02:16:50< celticminstrel> If you need help converting pseudo-code to C++ I can probably do that. 20180407 02:17:12< mattsc> The other question would be whether we should consider both problems simultaneously, so that we don’t have to do it twice. 20180407 02:17:35< mattsc> And then have a slightly inefficient method in 1.14. 20180407 02:17:52< celticminstrel> I guess that is indeed an option. 20180407 02:17:54< mattsc> This is just a linear operation fo a few simply math equations over a recall list. 20180407 02:18:07< mattsc> It’s not like this is going to take a lot of computation time, even in the worst case. 20180407 02:18:27< celticminstrel> I guess the recall filter is a "match this unit", not "find matches on map", so it should be fairly fast even when present. 20180407 02:18:45< celticminstrel> Iterating over advancements wouldn't really be slow because it's a very small list. 20180407 02:18:58< mattsc> Okay. 20180407 02:19:13< celticminstrel> So yeah I guess the contents of that loop isn't too inefficient, so it could be okay to just call it twice... 20180407 02:19:31< celticminstrel> Or I guess, several times? I'm not quite clear on what your plan was there. 20180407 02:19:52< celticminstrel> I think the proper fix would require changes to evaluation in addition to execution. The evaluation needs to consider the possibility of a recall too. 20180407 02:19:56< mattsc> I _think_ that it should be possible to get away with twice. 20180407 02:20:32< celticminstrel> Any comment on this? https://github.com/wesnoth/wesnoth/pull/2793/commits/731c798d0fc12fda4ce9c14f8610692aaccaf930 20180407 02:20:54< mattsc> Well, let me look at it on more carefully and come up with a conceptual solution. I’ll get back to you. 20180407 02:21:15< celticminstrel> FTR, IMO we should encourage using ai_algorithm rather than directly including one of the config files defining an AI. 20180407 02:21:41< mattsc> That commit should be fine. It’s not necessary to include that any more. 20180407 02:22:09< mattsc> Yeah, that (using ai_algorithm) is a good idea. 20180407 02:22:28< mattsc> I think the available algorithms are hard-coded though? Or did you change that? 20180407 02:22:31< celticminstrel> Might want to move the dev AIs out of their own namespace for that though. 20180407 02:22:40< celticminstrel> I did change what ai_algorithm does. 20180407 02:22:51< mattsc> Okay. 20180407 02:23:07< celticminstrel> IIRC, it previously selected amongst the registered C++ AI algorithms. 20180407 02:23:33< celticminstrel> But by 1.13.x, the only remaining C++ AI algorithms were the RCA algorithm and the idle AI IIRC. 20180407 02:24:00< celticminstrel> So now, it instead selects an [ai] tag by ID (from those listed in [ais], at the very least). 20180407 02:24:17< celticminstrel> (I haven't tested whether it can also match AIs defined in an active [era] or [modification].) 20180407 02:24:59< celticminstrel> A bunch of places used grouping=none instead of grouping=no for some reason. 20180407 02:25:17< celticminstrel> Pretty sure grouping=no is correct, based on a quick grep of the source. 20180407 02:25:18< mattsc> Okay, that sounds good. But it would have to be able to find user-defined AIs also. 20180407 02:25:45< celticminstrel> Ideally it would find any defined in the active era and any active modifications. 20180407 02:25:55< celticminstrel> And any that an addon placed in their own [ais] tag. 20180407 02:26:02< celticminstrel> But I haven't tested to see whether it does. 20180407 02:26:12< mattsc> AiWML says grouping=no, which is also consistent with my recollection 20180407 02:26:19< celticminstrel> So that might be something to look at in 1.15. 20180407 02:26:52< mattsc> Yeah. I think that would be a more elegant solution. 20180407 02:27:06< celticminstrel> A bunch of scenarios also use villages_per_scout=0.0 20180407 02:27:11< celticminstrel> Technically fine, but... 20180407 02:27:31< celticminstrel> villages_per_scout is an integer. 20180407 02:27:39< mattsc> Right. 20180407 02:27:50< mattsc> Which technically it wouldn’t have to be, but it is. 20180407 02:28:04< celticminstrel> I mean, I don't see how it would make sense as a real number. 20180407 02:28:28< celticminstrel> One campaign used [protect_area], was that ever a thing? 20180407 02:28:51< mattsc> It might have been. 20180407 02:28:51< celticminstrel> I changed it to [protect_location] (could've gone all the way to [goal] but went with the change that affected less lines). 20180407 02:29:08< celticminstrel> https://github.com/wesnoth/wesnoth/pull/2793/commits/c72b0d12d7ab43cdcfe1e7c041b71f9d566f5aa0#diff-24aaaa4e3e0df80a7c8ed921ede2dd66R207 20180407 02:29:10< mattsc> There were [protect_*] tags, but those were already deprecated by the time I started. 20180407 02:29:34< celticminstrel> The [protect_*] tags are supported in the 1.14 branch still. They're still deprecated too. 20180407 02:29:50< mattsc> Oh, area … 20180407 02:29:57< mattsc> Yeah, that probably did not exist. 20180407 02:30:12< celticminstrel> Though come to think of it, I don't think there's an actual deprecation message when they're encountered, maybe I should work on that at some point. 20180407 02:31:12< celticminstrel> ...and then there was someone who just dropped a [goal] straight in [side] for some reason... (Though I think that's also one of the misuses of goals that you've mentioned before?) 20180407 02:31:16< celticminstrel> https://github.com/wesnoth/wesnoth/pull/2793/commits/16dd9d553587e0ebd795aa5f2ff572fa083d7378 20180407 02:31:40< mattsc> There was some pretty vocal resistance to actually deprecating [target] (and probably the [protect] tags), probably during 1.11. 20180407 02:31:41< celticminstrel> (The second time I encountered something like that, I wrapped the goal in #ifdef WIP) 20180407 02:31:50< mattsc> Too much work to update one’s add-ons … 20180407 02:32:05< celticminstrel> Well, with how it is now, you don't have to! 20180407 02:32:13< celticminstrel> You can continue using the deprecated functionality indefinitely, probably. 20180407 02:33:04< mattsc> Right. Just telling you why they weren’t deprecated a long time ago. 20180407 02:33:16< mattsc> As for that [goal], it depends what the purpose is. 20180407 02:33:30< celticminstrel> A long time ago when they weren't deprecated, there wasn't a nice clear framework for deprecation. 20180407 02:33:35< mattsc> If you want the AI units to head toward Delfador, that will do it. 20180407 02:34:03< mattsc> I commented on the comment that said that was there so that the AI would *kill* the unit. 20180407 02:34:16< irker393> wesnoth/wesnoth:master Iris Morelle 3967fd190e gui2/log_settings: Make log options tool AppVeyor: All builds passed 20180407 02:34:17< celticminstrel> They're an obvious level 1 deprecation, which is basically "please don't use this in new code but it'll continue to work for the foreseeably future". 20180407 02:34:42< celticminstrel> Yeah, I have no idea what the intent was with that goal. 20180407 02:35:35< mattsc> It’s probably okay. It will make all the pillagers head toward Delf. 20180407 02:35:50< mattsc> And since he’s a high-value unit, they’ll likely also attack him. 20180407 02:36:31< celticminstrel> I assume you have no idea why several UtBS scenarios included "snapshot=no" in them. 20180407 02:36:44< mattsc> Umm. Correct. 20180407 02:37:11< celticminstrel> Yeah, here's the other case of goals in wrong places: https://github.com/wesnoth/wesnoth/pull/2793/commits/1bd9a46f6016e7e27bd7e0de3a36e58296e29589#diff-2a7863feacc0b0874ceb473a0543f5a7R274 20180407 02:37:37< celticminstrel> Oh, that's the last commit, even. 20180407 02:37:46< mattsc> Okay, yeah, that does not work. 20180407 02:37:56< mattsc> And the comment on top of it is incorrect, of course. 20180407 02:38:02< celticminstrel> There's two of them too. 20180407 02:38:03< mattsc> That’s exactly what I mean. 20180407 02:38:50< mattsc> I think the original choice of calling that [target] was unfortunate. 20180407 02:39:09< mattsc> Because what you intuitively imagine for that is an attack target. 20180407 02:39:25< mattsc> [goal] is much better, but the damage was done. 20180407 02:40:06< mattsc> So I suggest that you change the wording of that comment also. 20180407 02:40:19< celticminstrel> So I might yet have a question or two regarding the AI scenarios, as well, but I haven't yet gotten around to looking over them. 20180407 02:40:28< celticminstrel> The AI test scenarios I mean. 20180407 02:40:35< mattsc> Okay. 20180407 02:40:36< celticminstrel> Sure, I guess I can change the comment too. 20180407 02:41:08< mattsc> I am going to be traveling for most of the next week, but I’ll be online on and off. And as you know, I do grep the logs for my name. 20180407 02:42:32< mattsc> The recall bug needs to be fixed before the 15th? 20180407 02:42:54< mattsc> I’ll see what I can do. 20180407 02:42:59< celticminstrel> Yeah, I believe so. 20180407 02:43:11< celticminstrel> If not though, it could go into 1.14.1. 20180407 02:43:30< mattsc> Right. 20180407 02:43:52< mattsc> It shouldn’t be all that much work just for the basic fix. 20180407 02:45:03< mattsc> Just that I only have 2 days that I’ll be home before the 15th, and those aren’t exactly calm days. 20180407 02:45:29< mattsc> Anyways, it’s a lot less work than what you are doing. :) 20180407 02:47:35-!- gfgtdf [~chatzilla@x4e36316a.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.7.3/20180322140748]] 20180407 02:47:39< mattsc> Btw, if I did this in Lua, I’d set up a table during the first iteration that would look something like this: 20180407 02:48:14< mattsc> best_recall_by_type = { Spearman = id1, Bowman = id2, … } 20180407 02:48:37< mattsc> where id? are the respective ids of the highest_valued recall of that type 20180407 02:49:08< mattsc> And populated only with those units that are considered worth it. 20180407 02:49:44< mattsc> Then, there is no second iteration needed, you just pick the unit out of that table. 20180407 02:49:55< mattsc> I assume something like that is possible in C++ as well. 20180407 02:51:12< celticminstrel> std::map best_recall_by_type; 20180407 02:51:18< celticminstrel> ^ A variable that can hold that sort of data 20180407 02:51:32< mattsc> Cool. Thanks. 20180407 02:51:40< mattsc> I’ll still think about how to combine this with the second problem first though. 20180407 02:51:42< celticminstrel> Could use unordered_map instead of map, I guess. 20180407 02:51:49< mattsc> Okay. 20180407 02:51:54< celticminstrel> Probably doesn't matter that much though, 20180407 02:55:31-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity] 20180407 02:58:18< mattsc> Well, I guess for populating it it would have to look like: 20180407 02:58:50< mattsc> = { Spearman = { id = whatever_id, value = best_value }, … } 20180407 02:59:11< mattsc> But I’m sure I can figure that out. 20180407 02:59:34< celticminstrel> Well, that kinda complicates things a little, but probably the easiest way is to have two separate maps, one a type->ID map and the other a type->value map... but there might be some weird pitfalls in that approach, not sure. 20180407 02:59:57< celticminstrel> The other option is to define a struct that hols the ID and value, and use that as the map's value type. 20180407 02:59:57< mattsc> No, that should work. 20180407 03:00:11< celticminstrel> Or just use std::pair as the value type. 20180407 03:00:34< mattsc> Oh, okay, yeah, that would do. 20180407 03:01:06< celticminstrel> FTR in case it's not clear, "value type" refers to the second std::string in the map declaration. Maybe it's obvious, I dunno. 20180407 03:01:27< mattsc> I don’t know if it’s obvious, but that’s what I understood. 20180407 03:01:39< mattsc> So most likely it’s obvious, yes. ;) 20180407 03:01:54< mattsc> Okay, I got to disappear again. 20180407 03:01:58< mattsc> Thanks for all the help. 20180407 03:03:29< celticminstrel> Note that with pair you need to use .first and .second rather than .id and .value. 20180407 03:04:27< mattsc> Okay. 20180407 03:05:16< mattsc> I should really try to learn some more C++ sometime … 20180407 03:05:41< mattsc> I can read the code just fine, but writing it is a whole different story. 20180407 03:05:53< mattsc> Anyways, now I really am off. Thanks again. 20180407 03:06:49-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20180407 03:10:01< celticminstrel> So @Vultraz do you know whether village_gold or bonus are supported in [scenario] 20180407 03:10:19<+discordbot> I do not. 20180407 03:10:43< celticminstrel> The only instances in the source code are in WFL/Lua support, team.cpp, a GUI2 dialog, and here: https://github.com/wesnoth/wesnoth/blob/master/src/game_initialization/configure_engine.cpp#L195 20180407 03:10:50< celticminstrel> (That's for village_gold, haven't searched bonus yet.) 20180407 03:11:29< celticminstrel> There are zero instances of bonus in the source. 20180407 03:11:48< celticminstrel> (IIRC it's maybe something supported in [endlevel]? But I was only searching C++ source.) 20180407 03:16:32< celticminstrel> Haha, instant defeat on entering SoF7. 20180407 03:16:45< celticminstrel> Might be something to do with jumping to it via :choose_level though. 20180407 03:17:06< celticminstrel> Still, maybe someone should check it out/ 20180407 03:17:09< celticminstrel> ^? 20180407 03:25:35-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 03:25:40-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 03:30:19<+discordbot> please do 20180407 03:30:41< celticminstrel> Getting the same in 16_The_Kalian. 20180407 03:31:03< celticminstrel> I guess most likely both cases are the same - required hero units do not exist. 20180407 03:31:35<+discordbot> celmin: the one you linked is in [side]... 20180407 03:35:28< celticminstrel> Yeah, that's what I thought. 20180407 03:41:07<+discordbot> i mean... 20180407 03:41:08<+discordbot> for(config& side : scenario.child_range("side")) { 20180407 03:41:13<+discordbot> 😐 20180407 03:48:33< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes ed9ab7e48937 / data/campaigns/The_Rise_Of_Wesnoth/scenarios/ (21_The_Plan.cfg 23_Epilogue.cfg): TRoW: Fix a couple of issues found by the schema validator https://github.com/wesnoth/wesnoth/commit/ed9ab7e48937d6f881f4788ee3f555c988a4708b 20180407 03:48:35< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 652f3f3d3bdc / data/campaigns/Under_the_Burning_Suns/scenarios/ (05_A_Subterranean_Struggle.cfg 06b_In_the_Domain_of_Dwarves.cfg): fixup! UtBS: Fix a few more issues found by the schema validator https://github.com/wesnoth/wesnoth/commit/652f3f3d3bdc615622a48467cd77f2ba4f2e8d3d 20180407 03:48:37< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 8f1beea7b8d4 / data/campaigns/Heir_To_The_Throne/ (scenarios/01_The_Elves_Besieged.cfg utils/intro.cfg): HTTT: Fix toplevel [story] https://github.com/wesnoth/wesnoth/commit/8f1beea7b8d484ca5aae9648c6bd3f89ce199e5e 20180407 03:48:39< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 9a8e73697776 / data/campaigns/The_Rise_Of_Wesnoth/scenarios/ (15 files): TRoW: [scenario] does not support bonus=yes https://github.com/wesnoth/wesnoth/commit/9a8e736977763b598ebcb62a1a81595cf0089ac8 20180407 03:48:41< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes e23afd4f601a / data/campaigns/The_Rise_Of_Wesnoth/ (scenarios/16_The_Kalian.cfg units/Kalian.cfg): TRoW: [unit]image= is not supported https://github.com/wesnoth/wesnoth/commit/e23afd4f601aac4c6904de5b25cf48964e812ed8 20180407 03:48:43< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 191f0c1121e8 / data/campaigns/ (2 files in 2 dirs): HTTT, SoF: Fix a couple more issues found by the schema validator https://github.com/wesnoth/wesnoth/commit/191f0c1121e8ed04fd4c9d395f86b2209556f28e 20180407 03:48:45< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 465db5628970 / data/core/ (editor/music.cfg macros/sound-utils.cfg): Fix erroneous [event] in [editor_music] https://github.com/wesnoth/wesnoth/commit/465db5628970501d1933a5ff0a6984dff92b51af 20180407 03:48:53< celticminstrel> All that's left now is LoW and tests. 20180407 03:51:49<+discordbot> \o/ 20180407 03:51:55<+discordbot> you're doing good work here 20180407 03:59:22< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 1208d224c1ef / data/core/macros/sound-utils.cfg: fixup! Fix erroneous [event] in [editor_music] https://github.com/wesnoth/wesnoth/commit/1208d224c1efba246bc2ab671ca45c1158b8c7f5 20180407 04:07:09< celticminstrel> BTW is LoW supposed to allow you to start at any scenario in MP? 20180407 04:07:24< celticminstrel> I would've thought you're supposed to be allowed to start only at the first scenario of each chapter. 20180407 04:16:35< celticminstrel> Ohhh it's because of --debug 20180407 04:16:40< celticminstrel> I'm not sure how to feel about that? 20180407 04:16:57< celticminstrel> I think it's a bit surprising, but on the other hand you're not really supposed to play with --debug anyway. 20180407 04:18:13< irker393> wesnoth/wesnoth:1.14 Iris Morelle 5debd77a65 gui2/log_settings: Make log options tool AppVeyor: All builds passed 20180407 04:25:54< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 9014495f004b / data/scenario-test.cfg: Test scenario: remove useless [object]silent=yes https://github.com/wesnoth/wesnoth/commit/9014495f004b819715e8378389fd2095895e59de 20180407 04:25:56< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 61518e1c56bd / data/scenario-test.cfg: Test scenario: Fix invalid [message] key https://github.com/wesnoth/wesnoth/commit/61518e1c56bd3ace5a36649fbc953e0336669de5 20180407 04:25:58< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 2e9385cfb878 / data/scenario-story.cfg: Story test scenario: Fix misplaced duplicate map data https://github.com/wesnoth/wesnoth/commit/2e9385cfb87872c76a895ea4a36785d406cdbea2 20180407 04:26:00< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes e0ef234433e3 / data/campaigns/Legend_of_Wesmere/scenarios/chapter4/13_News_from_the_Front.cfg: LoW: [scenario]snapshot= is not supported https://github.com/wesnoth/wesnoth/commit/e0ef234433e31b856c55a21f0db860af966a93bc 20180407 04:26:02< irker393> wesnoth: Celtic Minstrel wesnoth:schema_fixes 65e0f232b29d / data/campaigns/Legend_of_Wesmere/scenarios/chapter3/12_Revelations.cfg: LoW MP: Fix a few more issues found by the schema validator https://github.com/wesnoth/wesnoth/commit/65e0f232b29d5cadf01435f5fe9c68f0113a5cc5 20180407 04:27:00<+discordbot> if(scenario["allow_new_game"].to_bool(false) || is_first || game_config::debug ) { 20180407 04:32:00-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 04:32:05-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 04:36:11< celticminstrel> The schema branch now has a whopping 132 commits; schema_fixes has 49. 20180407 04:36:20< celticminstrel> (That's after rebasing with autosquash.) 20180407 04:36:49< celticminstrel> The schema branch could be compressed a little more, I suppose; some of the schema system commits could be combined. 20180407 04:37:51< celticminstrel> Anyway, all that's left now is tests, which I will do tomorrow, and then #2793 is done (but please do start reviewing it now, as everything other than tests, which is to say all the most important parts, is already done). 20180407 04:38:15< celticminstrel> And then after that I want to clean up the schema stuff more to make it generally usable, some improved error-checking, etc. 20180407 04:38:26< celticminstrel> Probably gonna add a whole family of command-line options. 20180407 04:39:11< celticminstrel> Stuff like --validate-core, --validate-addon=ADDON_NAME, --validate-schema=path/to/schema.cfg, --validate file schema, and maybe more. 20180407 04:41:06-!- gallaecio [~quassel@57.99.79.188.dynamic.jazztel.es] has joined #wesnoth-dev 20180407 05:00:21< celticminstrel> I just wrote a schema for the schema format. \o/ 20180407 05:01:37<+discordbot> What’s the difference between the branches again? 20180407 05:01:51< celticminstrel> Between schema and schema_fixes, or...? 20180407 05:02:04<+discordbot> Yes 20180407 05:02:21< celticminstrel> schema contains all the work on the schema and fixes all the issues found in core WML by the validator. 20180407 05:02:32< celticminstrel> schema_fixes just cherry-picks all the fixes so they can go into 1.14. 20180407 05:02:39< celticminstrel> The fixes in core WML I mean. 20180407 05:02:49< celticminstrel> I guess I should say mainline rather than core, since most are in campaigns. 20180407 05:03:28-!- celticminstrel is now known as celmin|sleep 20180407 05:04:57<+discordbot> So not everything in schema is in schema_fixes, but the reverse is true? 20180407 05:06:40< celmin|sleep> Nearly correct. 20180407 05:06:50< celmin|sleep> There is one commit in schema_fixes that is not in schema. 20180407 05:07:12< celmin|sleep> The fix for SUF not supporting alignment/usage. 20180407 05:07:20< celmin|sleep> Because in schema that support is added instead of changing NR. 20180407 05:07:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180407 05:08:42<+discordbot> ok 20180407 05:08:46<+discordbot> perectyl acceptable 20180407 05:08:50<+discordbot> perfectly 20180407 05:09:00< celmin|sleep> There are in fact a number of commits in schema that technically have nothing to do with the schema itself. 20180407 05:09:21< celmin|sleep> Like the aforementioned ones adding features to SUF. 20180407 05:09:28<+discordbot> as long as all the fixes you make are present in some way on both branches 20180407 05:09:33<+discordbot> that's all that matters 20180407 05:09:47< celmin|sleep> Yeah, it's just that one NR thing that's done differently on both sides. 20180407 05:48:14-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180407 05:59:14-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20180407 06:52:35<+discordbot> @Pentarctagon it works for me 20180407 06:52:50<+discordbot> and presumably many others. 20180407 06:54:14-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20180407 06:54:57<+discordbot> How can I debug this GUI problem? 20180407 06:54:58<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432070763735154688/unknown.png 20180407 06:55:20<+discordbot> --log-debug=gui/layout : 20180407 07:00:51<+discordbot> Try editing this line: https://github.com/wesnoth/wesnoth/blob/1.14/data/themes/default.cfg#L145 20180407 07:01:29<+discordbot> The +30 is the width of the HP box, and from what I can tell, it's the absolute width (e.g. +60 would double it to 60 pixels). 20180407 07:02:00<+discordbot> The theme format is documented here: https://www.wesnoth.org/forum/viewtopic.php?p=213708#p213708 20180407 07:08:06<+discordbot> I knew, where to edit it, but I didn't know, how it works 😄 20180407 07:10:14-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20180407 07:15:52< irker393> wesnoth: Sofartin wesnoth:1.14 d51ccbe5386e / data/themes/default.cfg: Fix HP label in longer languages https://github.com/wesnoth/wesnoth/commit/d51ccbe5386e5325c6ecfe6c46472f7a8113bfa7 20180407 07:16:34<+discordbot> @hrubymar10 please forward-port 20180407 07:16:56<+discordbot> Didn't you already rip out the GUI1 sidebar in master? 20180407 07:17:07<+discordbot> oh right 20180407 07:17:09<+discordbot> duh 20180407 07:17:13<+discordbot> @hrubymar10 nevermind 20180407 07:17:16<+discordbot> 😂 20180407 07:17:20<+discordbot> D 20180407 07:17:23<+discordbot> @jyrkive thanks for reminding me 20180407 07:20:44-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 07:20:45< travis-ci> Pentarctagon/wesnoth#390 (master - 83c0fc5 : Sofartin): The build has errored. 20180407 07:20:45< travis-ci> Build details : https://travis-ci.org/Pentarctagon/wesnoth/builds/363415680 20180407 07:20:45-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 07:21:44<+discordbot> @Vultraz , did you see this: https://github.com/wesnoth/wesnoth/issues/2825 : 20180407 07:22:10<+discordbot> I've been following it, yes 20180407 07:23:36<+discordbot> Is it time to include l10n-track inside macOS package? 20180407 07:24:04<+discordbot> @shadowm ^^? Any update on #2825? 20180407 07:33:52< irker393> wesnoth: Jyrki Vesterinen wesnoth:opengl 3e96a3aa2c35 / / (5 files in 3 dirs): Add a shader class https://github.com/wesnoth/wesnoth/commit/3e96a3aa2c357c539c255dbe8b431f6857f8170a 20180407 07:39:22<+discordbot> honestly I don't like this need to include l10n-track 20180407 07:39:34<+discordbot> maybe it should be changed to use a config file in data/ dir instead 20180407 07:39:52<+discordbot> like translation completion tracking 20180407 07:53:11<+discordbot> That's the purpose of the file loonycyborg, you're just debating semantics here. 20180407 07:53:37<+discordbot> Also, no, I have been busy with other stuff and we still have some time before RC 3. 20180407 07:55:39< irker393> wesnoth: Nils Kneuper wesnoth:1.14 a44ffcdeb748 / / (5 files in 4 dirs): updated French translation https://github.com/wesnoth/wesnoth/commit/a44ffcdeb748a04b9770af59e28e4baa4f6e1ba3 20180407 07:55:46< irker393> wesnoth: Nils Kneuper wesnoth:master 6be41f51a5ef / po/ (wesnoth-dw/fr.po wesnoth-lib/fr.po wesnoth/fr.po): updated French translation https://github.com/wesnoth/wesnoth/commit/6be41f51a5ef4714749b05577509cd3ebab24cdb 20180407 08:01:17-!- grzywacz [~karol@89-70-226-147.dynamic.chello.pl] has joined #wesnoth-dev 20180407 08:03:55<+discordbot> the point is, that it's not in data/ dir for some reason, so it won't be automatically bulk copied 20180407 08:04:17<+discordbot> not all my packaging methods involve using install target 20180407 08:04:28<+discordbot> so I assume it won't be used then right? 20180407 08:05:04<+discordbot> and fuzzy images will be always shown 20180407 08:06:33<+discordbot> Ugh. 20180407 08:06:35<+discordbot> You're difficult. 20180407 08:07:00<+discordbot> I've not decided yet because I need to actually take a closer look at the images situation first, like I said in the ticket. 20180407 08:07:41<+discordbot> Also, you need to copy stuff from the root regardless (README.md, changelog.md), so your excuse is flimsy at best. 20180407 08:07:51<+discordbot> It just means you get to add another rule to your scripts. 20180407 08:08:48<+discordbot> but I have a lot of scripts to update. Adding new files shouldn't be done lightly because it's another chance for me to forget to add them somewhere 20180407 08:08:57<+discordbot> so please consider changing this requirement 20180407 08:09:18<+discordbot> You're basically admitting to not having full knowledge of your own packaging scripts. 20180407 08:10:07<+discordbot> Errare humanum est 20180407 08:10:08<+discordbot> The worst-case scenario isn't terribly disruptive either (as explained in the ticket). It just means that Wesnoth loads everything and outputs that warning message once during a session. 20180407 08:10:37<+discordbot> Telling me this in Latin doesn't exactly make my points any less valid. 20180407 08:12:20-!- grzywacz [~karol@89-70-226-147.dynamic.chello.pl] has quit [Remote host closed the connection] 20180407 08:13:12<+discordbot> If you prefer the file to be in data/, I'm sure that can be easily arranged, but it doesn't make a load of sense for the tool because the tool needs to work with both data/ and images/. 20180407 08:14:14<+discordbot> If you prefer generating a WML file with the required table... that's going to delay the ticket until after 1.14.0, that's for sure. 20180407 08:14:43<+discordbot> Because that requires adding new WML syntax. 20180407 08:15:13<+discordbot> if extra installation step for l10ntrack will be needed you'll need to communicate it to all packagers 20180407 08:15:13<+discordbot> (And WML is not the most efficient thing ever devised by humanity, so I'm not too happy about that option either.) 20180407 08:15:15<+discordbot> not just me 20180407 08:15:30<+discordbot> Yes loonycyborg, this applies for literally all packaging changes. 20180407 08:15:30< irker393> wesnoth/wesnoth:1.14 stevecotton 36a78d8856 LoW S09 Lua cleanup: fix a line that was AppVeyor: All builds passed 20180407 08:15:57<+discordbot> It's amazing how you want to make a big deal out of this before I even decide if it's worth it. 20180407 08:16:13<+discordbot> I don't make a big deal of it 20180407 08:16:20<+discordbot> And that you opted to be completely accepting of the proposal in the ticket only to come to Discord to argue about it. 20180407 08:16:53<+discordbot> Yes, you are, because the whole argument here boils down to "this changes things for packagers". 20180407 08:17:38<+discordbot> Well I can make it work in any way 20180407 08:18:45<+discordbot> I could save myself the time and effort of dealing with an orphaned piece of machinery to genearte the file, as well as with having this debate right now by just closing the issue and letting Wesnoth continue assuming all localised images to be up-to-date. 20180407 08:19:17<+discordbot> There certainly are better things I can do with my time than vouching for a feature I did not write and which fell victim to the bus factor issue I've been warning you all about for 8 years. 20180407 08:19:59<+discordbot> hmm 20180407 08:20:04<+discordbot> That hour and a half I spent researching the history behind it I could've spent dealing with actual work Vultraz needs me to do for 1.14. 20180407 08:20:05<+discordbot> but at current stage 20180407 08:20:15<+discordbot> is copying l10ntrack is needed? 20180407 08:20:15<+discordbot> Do you see why this is frustrating now? 20180407 08:20:42<+discordbot> Like I never did anything about that before 20180407 08:20:44<+discordbot> You shouldn't be asking that question if you took the time to read that text wall I posted before coming to argue against the proposal here. 20180407 08:20:54<+discordbot> people told me that it would be handled automatically 20180407 08:21:00<+discordbot> and copying data/ would be enough 20180407 08:21:12<+discordbot> I don't argue against it 20180407 08:21:16<+discordbot> You copy more than just data/ righ tnow. 20180407 08:21:37<+discordbot> Unless your builds are all failing at an early stage of startup because a required font in fonts/ cannot be loaded. 20180407 08:22:26<+discordbot> As I already said, it's not the only solitary file on the data dir/repo root, and odds are it's not going to be the last one either. 20180407 08:24:52<+discordbot> Atl for Czech, images aren't so out-to-date btw 20180407 08:25:01<+discordbot> My take is that the status quo is fine. 20180407 08:25:21<+discordbot> The worse that happens is that the translated Help images use the old font 20180407 08:25:43<+discordbot> So in practical terms this debate has taken up more time than anyone should be spending on it 20180407 08:25:44<+discordbot> Or outdated UI (the Recruit dialog changed in 1.13.x, for starters). 20180407 08:26:06<+discordbot> I wish someone had told me that upfront though. 20180407 08:26:14<+discordbot> Old Recruit: 20180407 08:26:14<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432093732335255553/recruit.png 20180407 08:26:23<+discordbot> New one: 20180407 08:26:23<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432093770495033344/recruit.png 20180407 08:26:35<+discordbot> Not so big difference IMO 20180407 08:26:47<+discordbot> Feel free to mark the issue as Invalid or assign it to someone else with more time in their hands. 20180407 08:31:47-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 08:31:52-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 08:32:28<+discordbot> Well again: I don't feel, that it is necessary to include l10n-track inside packages in any way. Let's remove this requirement from src and done. But when will shadowm have required time, then she will add the script which handles images into our repo and then there will be 2 options: 1) Someone for will have to run this script once per year or half and will have to inform translation teams if their image became fuzzy 2) Make 20180407 08:32:29<+discordbot> it as part of po-update, so this script will be used more often and again, who will run po-update will also have responsibility for informing lang teams (I hope that I wrote smt understandable 😄 ) 20180407 08:33:17< irker393> wesnoth: Charles Dang wesnoth:1.14 1cb76697134d / src/image.cpp: Image: handle case where l10n-track file is not present https://github.com/wesnoth/wesnoth/commit/1cb76697134da175e4ed12d718d06151bd6966e3 20180407 08:33:21<+discordbot> The plan was to go for 2, except as part of pot-update rather than po-update. I guess either one works, really. 20180407 08:33:21-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 08:33:23< travis-ci> wesnoth/wesnoth#17488 (master - 6be41f5 : Nils Kneuper): The build is still failing. 20180407 08:33:23< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363430369 20180407 08:33:23-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 08:34:06-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev 20180407 08:34:25<+discordbot> Making it an extra manual step for anyone was never part of the plan since that is precisely what the status quo was and what led us to the present situation after yet another dev monopolizing maintenance tasks decided to go AWOL. 20180407 08:34:36< irker393> wesnoth: Charles Dang wesnoth:master d786fd6622c6 / src/image.cpp: Image: handle case where l10n-track file is not present https://github.com/wesnoth/wesnoth/commit/d786fd6622c64660c88ced38c1a442eb9a7389c7 20180407 08:35:06<+discordbot> @hrubymar10 warning should be gone 20180407 08:38:22<+discordbot> Yes @Vultraz , this will remove warning/error but also kill requirement of l10n-track at all, because now I don't have to fix this warning by including l10n-track inside package. And it is not about me. Other packagers with 0 knowledge of this problem won't even know that this file exist. Ofc scons and cmake will handle it for them but still... 20180407 08:40:33<+discordbot> I'll include l10n-track inside macOS package and update Czech images. Please somebody send mail to translators with short summary and conclusion that if their language is mentioned in @shadowm 's diff, then they should update their images too 20180407 08:47:33< irker393> wesnoth: Sofartin wesnoth:1.14 eacd2a4cd17f / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Add l10n-track to macOS Package - #2825 https://github.com/wesnoth/wesnoth/commit/eacd2a4cd17fa5807b8c087e0cd64de51c8446a8 20180407 08:54:38< irker393> wesnoth: Sofartin wesnoth:master d27e359d6eaf / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Add l10n-track to macOS Package - #2825 https://github.com/wesnoth/wesnoth/commit/d27e359d6eafe840d7c8fb3f75483a6b9ab29f3b 20180407 09:00:33-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 09:00:42-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 09:02:27-!- atarocch [~atarocch@37.176.80.255] has joined #wesnoth-dev 20180407 09:07:56-!- travis-ci [~travis-ci@ec2-54-196-99-208.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 09:07:57< travis-ci> wesnoth/wesnoth#17490 (master - d786fd6 : Charles Dang): The build is still failing. 20180407 09:07:57< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363436977 20180407 09:07:57-!- travis-ci [~travis-ci@ec2-54-196-99-208.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 09:11:23<+discordbot> not sure why that's failing 🤔 20180407 09:12:45<+discordbot> the call is correct, and it includes the file 20180407 09:21:29<+discordbot> @shadowm why do you think that there even is an updated version of that script? You said you wanted to look for it. 20180407 09:22:28<+discordbot> Will you drop it already? 20180407 09:22:49<+discordbot> Well that was so long time ago, I forgot most of that stuff already 20180407 09:23:01<+discordbot> It wasn't even your business at the time. 20180407 09:23:14<+discordbot> did that gsoc student plant to use it in other projects too? 20180407 09:23:17<+discordbot> The bottomline is I'm not dealing with this anymore, so stop bothering me about it. 20180407 09:23:23<+discordbot> @Vultraz Note that it's the C++ unit tests which fail to link, not the game itself. 20180407 09:23:25<+discordbot> What GSoC student? 20180407 09:23:39<+discordbot> who made image localization 20180407 09:23:45<+discordbot> iirc it was part of gsoc 20180407 09:24:12<+discordbot> Pretty sure it wasn't. 20180407 09:24:34<+discordbot> A project that small wouldn't have met the requirements for GSoC slots at the time. 20180407 09:25:49<+discordbot> Also, he was a translator for us, and a KDE contributor (an attentive eye will notice his name is in the script itself). 20180407 09:26:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180407 09:34:17<+discordbot> yes I noticed it 20180407 09:34:30<+discordbot> Maybe it's used in kde somewhere? 20180407 09:34:54<+discordbot> Or most likely they have better things to do 20180407 09:35:30<+discordbot> the thing is I might be interested in integrating that stuff to pot-update myself later 20180407 09:37:42<+discordbot> But I didn't know enough about our translation pipeline besides what was in pot-update and related targets 20180407 09:42:15-!- travis-ci [~travis-ci@ec2-54-197-124-80.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 09:42:16< travis-ci> wesnoth/wesnoth#17492 (master - d27e359 : Sofartin): The build is still failing. 20180407 09:42:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363440163 20180407 09:42:16-!- travis-ci [~travis-ci@ec2-54-197-124-80.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 09:52:52-!- atarocch [~atarocch@37.176.80.255] has quit [Ping timeout: 256 seconds] 20180407 09:54:51< irker393> wesnoth/wesnoth:1.14 Celtic Minstrel deeafd6623 fixup incorrect function AppVeyor: All builds passed 20180407 10:01:25< irker393> wesnoth: newfrenchy83 wesnoth:master 2c65dff722f1 / projectfiles/CodeBlocks/wesnoth.cbp: Update wesnoth.cbp https://github.com/wesnoth/wesnoth/commit/2c65dff722f16d4c0a0892941c05062ccff59966 20180407 10:01:27< irker393> wesnoth: Martin Hrubý wesnoth:master 165349104d98 / projectfiles/CodeBlocks/wesnoth.cbp: Merge pull request #2840 from newfrenchy83/patch-11 https://github.com/wesnoth/wesnoth/commit/165349104d98f853e5391661a6169dfb23a935de 20180407 10:04:16< irker393> wesnoth: newfrenchy83 wesnoth:1.14 9ce33b8a945f / projectfiles/CodeBlocks/wesnoth.cbp: Update wesnoth.cbp - #2840 https://github.com/wesnoth/wesnoth/commit/9ce33b8a945f9e3a337b778d4989995db2a8bc96 20180407 10:06:24<+discordbot> oh... f*** 😄 20180407 10:08:02< irker393> wesnoth: Sofartin wesnoth:1.14 ec761f88d6fa / projectfiles/CodeBlocks/wesnoth.cbp: Revert "Update wesnoth.cbp - #2840" https://github.com/wesnoth/wesnoth/commit/ec761f88d6fa60282366e22a4c21c83c4857a45f 20180407 10:08:24<+discordbot> Sorry 😧 20180407 10:08:50-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 10:08:51< travis-ci> wesnoth/wesnoth#17495 (1.14 - 9ce33b8 : newfrenchy83): The build was canceled. 20180407 10:08:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363452935 20180407 10:08:51-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 10:16:00< irker393> wesnoth: Jyrki Vesterinen wesnoth:master e7ed3cbbb9ad / src/CMakeLists.txt: Fix linking of C++ unit tests with CMake https://github.com/wesnoth/wesnoth/commit/e7ed3cbbb9ad474f2f89137fceece946eaa95e81 20180407 10:16:27< irker393> wesnoth: loonycyborg wesnoth:1.14 823d37c37a12 / packaging/windows/Wesnoth.nsi.in: nsi script: update expected extensions of changelogs https://github.com/wesnoth/wesnoth/commit/823d37c37a12cbcef7c089eefaf9694fde753ab0 20180407 10:19:29-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 10:19:30< travis-ci> wesnoth/wesnoth#17494 (master - 1653491 : Martin Hrubý): The build is still failing. 20180407 10:19:30< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363452189 20180407 10:19:30-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 10:20:37< irker393> wesnoth: loonycyborg wesnoth:master 0e5463fd818b / packaging/windows/Wesnoth.nsi.in: nsi script: update expected extensions of changelogs https://github.com/wesnoth/wesnoth/commit/0e5463fd818b8cce388142c3ea949a289f6340ef 20180407 10:29:38<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432124787582566400/wesnoth_new_UI_mockup_V2.JPG 20180407 10:51:10<+discordbot> I still prefer it more under minimap on right 20180407 10:51:20<+discordbot> And make ToD transparent 20180407 10:51:34<+discordbot> (if possible) 20180407 10:53:11<+discordbot> I still very much dislike the font because it looks so thin. Weren't there some mock-ups some time ago with bigger fonts? 20180407 10:55:24-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 10:55:25< travis-ci> wesnoth/wesnoth#17498 (1.14 - 823d37c : loonycyborg): The build passed. 20180407 10:55:25< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363454766 20180407 10:55:25-!- travis-ci [~travis-ci@ec2-54-224-203-71.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 10:56:51<+discordbot> When you are talking about fonts/texts, there will be problem with MP, XP, MP... Translations... For example in czech: Životů, Zkušenosti, Pohyb 20180407 10:58:57< irker393> wesnoth/wesnoth:1.14 Celtic Minstrel f92abba0b7 fixup incorrect function AppVeyor: All builds passed 20180407 11:10:07-!- travis-ci [~travis-ci@ec2-54-196-99-208.compute-1.amazonaws.com] has joined #wesnoth-dev 20180407 11:10:08< travis-ci> wesnoth/wesnoth#17499 (master - 0e5463f : loonycyborg): The build was fixed. 20180407 11:10:08< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/363455357 20180407 11:10:08-!- travis-ci [~travis-ci@ec2-54-196-99-208.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180407 11:22:40<+discordbot> @Kwandulin the font you're seeing in the mockup is Sans, not Lato (which we use in the game) 20180407 12:05:25-!- grzywacz [~karol@wesnoth/developer/grzywacz] has joined #wesnoth-dev 20180407 12:06:17< grzywacz> Looks like I've managed to create an invalid map with the 1.13.14 editor: 20180407 14:04:41 error general: The game map could not be loaded: Illegal tile in map: (Xuq) 'Xuq' 20180407 12:07:28-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Ping timeout: 260 seconds] 20180407 12:07:39-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 12:13:21< vn971> grzywacz: you both created it and open it in same wesnoth version, right? 20180407 12:13:34< grzywacz> vn971: correct 20180407 12:14:02< grzywacz> vn971: it loads in the editor, but fails to start the scenario 20180407 12:14:22< grzywacz> same version for sure, because I switch between the game and the editor from the titlescreen 20180407 12:24:06<+discordbot> must mean you're not loading it in game 20180407 12:24:08<+discordbot> only in the editor 20180407 12:26:07< grzywacz> Not sure what you mean. It loads/saves in the editor, but game fails to load a campaign scenario with this map. 20180407 12:27:52<+discordbot> well, Xuq is not a core terrain 20180407 12:28:14<+discordbot> so it must be a custom one 20180407 12:28:48< grzywacz> ohhh 20180407 12:28:51<+discordbot> if you're only loading the terrain definition in the editor (ie, it's inside #ifdef EDITOR), then it'snot loaded in-game 20180407 12:28:59< grzywacz> Where did this come from... 20180407 12:29:54< grzywacz> Something's in a weird state, because that's the only terrain code for Cave Wall I see in my editor palette. 20180407 12:30:19< grzywacz> And I haven't done any custom terrains myself. 20180407 12:31:07<+discordbot> The only cave wall codes are Xu, Xue, and Xuc 20180407 12:31:47< grzywacz> data/add-ons/To_Lands_Unknown/macros/terrain_base.cfg 20180407 12:31:50< grzywacz> That evil addon... 20180407 12:31:55<+discordbot> do you have any addons installed that might define Xuq as a cusom t 20180407 12:32:00<+discordbot> yeah that 20180407 12:32:14<+discordbot> why is TLU evil? 20180407 12:34:14< grzywacz> Because it overwrote on Xu cave wall in the editor. 20180407 12:34:19< grzywacz> I didn't have two to choose from. 20180407 12:34:26< grzywacz> s/on/the/ 20180407 12:35:16<+discordbot> file a bug with inferno8 20180407 12:35:24<+discordbot> he shouldn't be overwriting core terrains... 20180407 12:37:28<+discordbot> ooh, that Wesnoth build above looks good 20180407 12:37:32<+discordbot> though the UI needs some work 20180407 12:37:48<+discordbot> also that DiD map looks different than i remembrer, did it get changed or something? 20180407 12:38:02<+discordbot> since im pretty sure that's Malin Keshar as the hero, there at the base 20180407 12:38:28< grzywacz> Vultraz: well, yeah, should I catch him on forums or something? Or is it a bug in the game that it allows overwriting core terrains from addons? 20180407 12:38:32<+discordbot> (i did finish DiD a few months ago so i have good memory) 20180407 12:38:40<+discordbot> @VYNLT it's a redrawn version I have locally but have not yet committed 20180407 12:38:47<+discordbot> looks really good tbh 20180407 12:38:48<+discordbot> @zookeeper objects to the paved roads 20180407 12:38:49<+discordbot> much more detailed 20180407 12:39:01<+discordbot> though i dont remember the bridge 20180407 12:39:05<+discordbot> is it Return To Parthyn? 20180407 12:39:08<+discordbot> or the first mission? 20180407 12:39:51<+discordbot> both would use the same map 20180407 12:39:54<+discordbot> oh 20180407 12:40:07<+discordbot> still, i feel like the bridge could make the mission easier, since the units dont get downgrades from the water terrain 20180407 12:40:28< irker393> wesnoth/wesnoth:master stevecotton 2b406a37ec LoW S09 Lua cleanup: fix a line that was AppVeyor: All builds passed 20180407 12:40:54<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432157823418368021/w.jpg 20180407 12:41:22<+discordbot> If I'm being honest, about 90% of mainline maps need to be redrawn more detailed at some point 20180407 12:41:27<+discordbot> yeah i agree 20180407 12:41:42<+discordbot> well if you want to keep the bridges at least turn that stone bridge into another wooden bridge 20180407 12:41:50<+discordbot> Parthyn is a village, not a city, so yeah 20180407 12:41:59<+discordbot> I did this a few months back since I was testing so much in DiD and was so sick of looking at that bland map 😛 20180407 12:42:11<+discordbot> even though its current iteration as you see was "improved" by myself 20180407 12:42:26<+discordbot> yeah you gotta submit this 20180407 12:42:28<+discordbot> its good 20180407 12:42:45<+discordbot> that Malin sprite also looks different, or is it just me? 20180407 12:43:00<+discordbot> did he get an update too? since i remember him looking more like Konrad's head on a mage body with a sword 20180407 12:43:15<+discordbot> I don't think so 20180407 12:43:23<+discordbot> I'll commit it at some point 20180407 12:46:21<+discordbot> still last time i played DiD i do think Malin did also have a different sprite 20180407 12:46:26<+discordbot> actually lemme check how he looks in my version of the game 20180407 12:47:44<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432159532358303757/unknown.png 20180407 12:47:45<+discordbot> yeah same sprite as yours i think 20180407 12:47:54<+discordbot> so i dont think he got changed 20180407 12:48:58<+discordbot> bllaaaanddddmap 20180407 12:48:59<+discordbot> 😛 20180407 12:49:55<+discordbot> mhm 20180407 12:50:03<+discordbot> i think the camp looks fits better for a village like Parthyn 20180407 12:50:13<+discordbot> i mean they do defend themselves from orcs but they do get casualities, so they're not professionals 20180407 12:50:21<+discordbot> i mean, Malin's father dies at the beginning of the campaign 20180407 12:50:39<+discordbot> and his sister gets wounded as mentioned at the end of the mission if you don't use undead units 20180407 12:50:55<+discordbot> i guess I'll revert the road to dirt 20180407 12:50:59<+discordbot> and make it a wooden bridge 20180407 12:51:31<+discordbot> yeah 20180407 12:51:41<+discordbot> Parthyn can defend itself from orcs but not without casualities 20180407 12:51:53<+discordbot> though in Return To Parthyn, Malin's sister does have powerful forces 20180407 12:52:02<+discordbot> if i remember correctly she can train Royal Guards (as in, max tier units) 20180407 12:52:06<+discordbot> not sure about it though 20180407 12:54:08<+discordbot> heh, I just realized the closed wooden door terrain could conceivably double for a wooden wall for someone's scenario 🤔 20180407 12:54:40<+discordbot> odd scale, sometimes have we 20180407 12:55:51<+discordbot> yeah 20180407 12:55:58<+discordbot> tbh though cities can look good in the game even when small 20180407 12:56:03<+discordbot> look at Elsenfar from Heir To The Throne 20180407 12:56:06<+discordbot> that one is a decent size 20180407 12:57:09<+discordbot> though Weldyn doesnt look good imo in the same campaign 20180407 12:57:15<+discordbot> (in the last mission where you fight Asheviere and her goons) 20180407 12:57:22<+discordbot> its just kinda...small and simple 20180407 12:57:34<+discordbot> i mean the city does have 3 other enemy forts around it but still 20180407 12:58:29<+discordbot> yeah 20180407 12:58:43<+discordbot> I think that map's been the same since it was first drawn 😛 20180407 12:58:50<+discordbot> 15 years ago 20180407 12:59:20<+discordbot> Weldyn really needs to look like a grander city tbh 20180407 12:59:25<+discordbot> i mean, it's where the FRIGGIN QUEEN OF WESNOTH IS 20180407 12:59:47<+discordbot> there's also Halstead from Liberty which, while big (which is appreciate), it looks a bit..empty 20180407 12:59:51<+discordbot> A city map for my campaign 20180407 12:59:52<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432162591574589440/SoD_E1S1.jpg 20180407 12:59:53<+discordbot> it doesnt really have much going for it 20180407 13:00:02<+discordbot> looks good for a city 20180407 13:00:09<+discordbot> i also made my own city map (though it also has outskirts) 20180407 13:00:12<+discordbot> lots of villages 20180407 13:01:16<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432162939688976384/unknown.png 20180407 13:01:18<+discordbot> it is a little basic but i do think i got the size right 20180407 13:01:48<+discordbot> as well as the contents (so it doesnt look as bland as Halstead) 20180407 13:02:03<+discordbot> (even though Halstead is a fortress, not really a city) 20180407 13:02:07<+discordbot> ....huh 20180407 13:02:27<+discordbot> ? 20180407 13:02:41<+discordbot> not exactly my style, but it's serviceable 20180407 13:02:50<+discordbot> yea, the shape is kinda basic 20180407 13:02:52<+discordbot> standard hexagon 20180407 13:03:07<+discordbot> im not exactly the best at adding details to maps 20180407 13:03:27<+discordbot> shadowm makes the absolute best maps 20180407 13:03:32<+discordbot> yeah, he's good 20180407 13:03:37<+discordbot> she* 20180407 13:03:46<+discordbot> i love the Chaos Empire fortress in that one mission from Invasion From The Unknown 20180407 13:03:52<+discordbot> towers that actually attack you, good scale, good content 20180407 13:03:57<+discordbot> and good outlying defenses 20180407 13:04:15<+discordbot> (its so good that i never managed to beat that mission lol, its just so damn hard) 20180407 13:04:31<+discordbot> i had a good army of Necromancers but i lost it thanks to the Shaxtals in the earlier missions, i swear, fuck the Shaxtals 20180407 13:04:35<+discordbot> oh, yeah, that's a hard one 20180407 13:04:46<+discordbot> did you play After the Storm? 20180407 13:04:51<+discordbot> just started it 20180407 13:04:54<+discordbot> so far so good 20180407 13:04:56<+discordbot> 👌 20180407 13:05:01<+discordbot> you're in for a treat 20180407 13:05:14<+discordbot> i did read what happens after that mission so yeah i will be familiar with the stuff in that one 20180407 13:05:15<+discordbot> it surpasses IftU in some regards 20180407 13:05:18<+discordbot> yeah i heard 20180407 13:05:36<+discordbot> one of the best parts is Malin imo 20180407 13:05:41<+discordbot> i love how he roasts everybody 20180407 13:05:48<+discordbot> indeed 20180407 13:06:02<+discordbot> he's pretty much the Obi-Wan Kenobi of the IFTU/ATS series 20180407 13:06:17<+discordbot> H E L L O T H E R E 20180407 13:06:54<+discordbot> honestly, stuff like IftU, AtS, and To Lands Unknown put mainline wesnoth to shame 20180407 13:07:13<+discordbot> true, at least most of the mainline ones 20180407 13:07:20<+discordbot> yeah 20180407 13:07:24<+discordbot> though concepts from those wouldnt exactly exist without a few of the mainline ones 20180407 13:07:33<+discordbot> we haven't really focused on the actual game content as much as we shoul 20180407 13:07:33<+discordbot> since ATS and IFTU are sequels to Under The Burning Suns 20180407 13:07:51<+discordbot> as well as, Descent Into Darkness considering Malin Keshar returns as a main character 20180407 13:07:53<+discordbot> i redid a bunch of maps for NR (first 8 or 9) in early 2015, though 20180407 13:07:57<+discordbot> ah 20180407 13:08:13<+discordbot> UtBS is getting an upgrade in 1.14 20180407 13:08:16<+discordbot> whow 20180407 13:08:17<+discordbot> whoa 20180407 13:08:19<+discordbot> rly 20180407 13:08:24<+discordbot> what is it gonna be about? 20180407 13:08:27<+discordbot> well 20180407 13:08:32<+discordbot> other than unique Quenoth Elves models 20180407 13:08:36<+discordbot> (which i heard rumors about) 20180407 13:08:47<+discordbot> it's mostly just the new quenoth 😛 20180407 13:08:50<+discordbot> oh 20180407 13:08:58<+discordbot> BU 20180407 13:08:59<+discordbot> T 20180407 13:09:01<+discordbot> hm 20180407 13:09:05<+discordbot> I'll be working on improving it for 1.16 20180407 13:09:09<+discordbot> ah 20180407 13:09:15<+discordbot> it deserves better 20180407 13:09:21<+discordbot> well i guess i will be replaying it once it gets enough additions, even though its gonna be hell 20180407 13:09:33<+discordbot> (i swear UTBS really kicked my ass in some missions, even if i managed to pass the campaign) 20180407 13:09:52<+discordbot> will be worth it to see the beautiful new portraits by LordBob that will accompany the new elves 20180407 13:09:55<+discordbot> it wasnt as bad as some of IFTU's missions but still, UTBS is a hard motherfucker 20180407 13:09:58<+discordbot> new portraits? oh boy 20180407 13:10:23<+discordbot> https://forums.wesnoth.org/viewtopic.php?p=625570#p625570 20180407 13:10:25<+discordbot> for example 20180407 13:10:26<+discordbot> (fyi i dont mean that the hard difficulty is a negative part, im just saying that it's hard 20180407 13:10:40<+discordbot> ooh i like the artstyle 20180407 13:10:46<+discordbot> very exotic and fitting 20180407 13:11:04<+discordbot> yeah, LordBob is an excellent artist 20180407 13:12:25<+discordbot> @loonycyborg and maybe @Vultraz , can you please comment here? https://github.com/wesnoth/wesnoth/pull/2841 20180407 13:12:48<+discordbot> @hrubymar10 as far as i can tell it is fine 20180407 13:13:19<+discordbot> I also thought but then @sevu came 😄 20180407 13:13:21<+discordbot> @VYNLT are you still playing on 1.12? 20180407 13:13:35<+discordbot> yeah 20180407 13:13:40<+discordbot> i was waiting for 1.13 to get stable 20180407 13:13:48<+discordbot> ah, so you haven't tried 1.13 yet? 20180407 13:13:51<+discordbot> no 20180407 13:13:55<+discordbot> im waiting for stable versions 20180407 13:14:06<+discordbot> well, right now it's in Release Candidate stage 20180407 13:14:15<+discordbot> so that means stable enough? 20180407 13:14:15<+discordbot> it's stable 20180407 13:14:17<+discordbot> yes 20180407 13:14:18<+discordbot> mkay 20180407 13:14:22<+discordbot> how do i port saves though? 20180407 13:14:30<+discordbot> and, well, content 20180407 13:14:45<+discordbot> actually, how do i update? 20180407 13:14:48<+discordbot> first time updating 20180407 13:14:59<+discordbot> just download a new version and install it separately 20180407 13:15:10<+discordbot> or you can wait until may and download it off Steam 20180407 13:15:13<+discordbot> alright so im downloading 1.13.13- 20180407 13:15:17<+discordbot> yeah i think i'll wait 20180407 13:15:37<+discordbot> for saves, you can just copy them to the new version's userdata folder 20180407 13:15:47<+discordbot> and as for content, there's a lot of API changes for UMC creators 20180407 13:15:57<+discordbot> oh 20180407 13:15:58<+discordbot> so its, better? 20180407 13:16:30<+discordbot> 1.14 will be 10x better than 1.12 20180407 13:16:51<+discordbot> mk 20180407 13:17:10<+discordbot> you'll need to keep 1.12 around if you want to play AtS though 20180407 13:17:16<+discordbot> shadowm hasn't ported it yet 20180407 13:17:18<+discordbot> mk 20180407 13:17:23<+discordbot> she's been busy 20180407 13:17:27<+discordbot> will wait for 1.14 then 20180407 13:19:33<+discordbot> @hrubymar10 I did build on 1.14 on linux, not on mac. 20180407 13:19:58<+discordbot> @hrubymar10 In case it matters... 20180407 13:24:45<+discordbot> I don't use this wesnoth.cbp project at all. I tried answer autor's question. 20180407 13:25:14-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20180407 13:25:43<+discordbot> But tbh your answer seems to be weird for me. Yes, it could be different on linux but I don't think that SDL2_image depends on libpng 20180407 13:26:05<+discordbot> that's whole story 😄 @sevu ^^ 20180407 13:26:35<+discordbot> I think it's a runtime dependency, i.e. that SDL_Image loads libpng at runtime to be able to decode/encode PNG images. 20180407 13:28:00<+discordbot> Is it? If is, then sorry for my mistake. 20180407 13:28:13<+discordbot> I see. Surprisingly master doesn't build for me without libpng, it fails while compiling complaining about pango then. 20180407 13:29:01<+discordbot> IIRC, the GTK+ stack (where Pango is) also needs libpng for something. 20180407 13:29:03<+discordbot> pango and cairo... I know that libpng is dependence of cairo. So that could be problem 20180407 13:31:05< celmin|sleep> Not that I mind discussing all that kind of thing, but this is the dev channel, wouldn't the main channel be better @VYNLT 20180407 13:33:44-!- celmin|sleep is now known as celticminstrel 20180407 13:37:15-!- octalot [~steve@178.115.128.204.wireless.dyn.drei.com] has joined #wesnoth-dev 20180407 13:42:06<+discordbot> oh sorry 20180407 14:09:24< irker393> wesnoth/wesnoth:1.14 Celtic Minstrel 4a672184f8 LoW MP: Fix a few more issues found by t AppVeyor: All builds passed 20180407 14:09:36< celticminstrel> \o/ 20180407 14:21:41<+discordbot> I tried fix it myself but I don't know how to do it correctly... 20180407 14:21:42<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432183187418906665/unknown.png 20180407 14:22:14<+discordbot> [add] [action] id=button-continue ref=show-what title= _ "Continue Play" tooltip= _ "Exit the replay mode and continue playing normally." items=replayexit rect="+15,=,+100,=" xanchor=fixed yanchor=fixed [/action] [/add] [add] [action] 20180407 14:22:15<+discordbot> id=skip-animation ref=button-continue type=checkbox title= _ "Skip animation" items=replayskipanimation rect="+15,=+2,+80,+10" xanchor=fixed yanchor=fixed [/action] [/add] 20180407 14:22:21< celticminstrel> I think the ref for the checkbox needs changing. 20180407 14:22:37< celticminstrel> It should reference the button currently labelled pokrasomething. 20180407 14:22:54<+discordbot> I tried make button-continue wider, I made it +150 20180407 14:23:03<+discordbot> It worked but in English it was ugly 😄 20180407 14:23:10< celticminstrel> So pokrasomething is continue? 20180407 14:23:19<+discordbot> yes 😃 20180407 14:23:39< celticminstrel> Okay, then the checkbox needs ref=button-continue... but isn't that what it has already? 20180407 14:23:59< celticminstrel> I can't read your pasted code very well though. 20180407 14:24:00<+discordbot> yes, it does. 20180407 14:24:13< celticminstrel> The checkbox is "skip animation"? 20180407 14:24:13<+discordbot> I can create github issue for you, ok? 20180407 14:24:41<+discordbot> yes, checkbox is skip-animation 20180407 14:24:47<+discordbot> like "skip animation" 20180407 14:25:17< celticminstrel> So the rect is +15,=+2,+80,+10... 20180407 14:26:05< celticminstrel> So +15 means "15px to the right of button-continue's right edge" 20180407 14:26:05<+discordbot> yes, id skip-animation have this rect 20180407 14:26:18< celticminstrel> =+2 means "2px below button-continue's top edge" 20180407 14:26:34<+discordbot> you don't get problem correctly 20180407 14:27:09< celticminstrel> +80 means "80px wide" and +10 means "10px tall". 20180407 14:27:11<+discordbot> problem is, that button-continue has max 100px wide which is not ok for Czech. Czech needs 150px. 20180407 14:27:51< celticminstrel> FTR I don't think it's possible to make the width depend on the content. 20180407 14:28:11<+discordbot> checkbox 'behind' button-continue works fine with button-continue's max 100px. Problem is that button-continue is wider in real 20180407 14:32:20< celticminstrel> I don't know what you mean by "wider in real". 20180407 14:32:51< celticminstrel> Are you saying that the button does expand to fit its content, but its rect for reference is unchanged when this happens? 20180407 14:37:04<+discordbot> I tried to say that while using czech, button-continue is wider then 100px 20180407 14:37:20<+discordbot> it is wide for example 148px 20180407 14:37:43< celticminstrel> Does making it wider in the theme work? 20180407 14:38:09< celticminstrel> IMO that would be acceptable even if it looks a little weird for English (though ask @Vultraz too). 20180407 14:38:32<+discordbot> yes, when I make button-continue 150px wide, then it works for czech 20180407 14:38:53<+discordbot> but in english (do you see images from discord in IRC?) 20180407 14:39:11< celticminstrel> I see them as a clickable link. 20180407 14:40:09<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/432187832870895617/unknown.png 20180407 14:41:06< celticminstrel> Oh, that's not as bad as I was imagining. 20180407 14:41:47<+discordbot> welp 😄 I don't like tbh 😄 @Vultraz ? ^^ 20180407 14:42:04<+discordbot> or you could move the checkbox before the buttons 20180407 14:42:20< celticminstrel> IMO that would look weird too, but yes it's an option. 20180407 14:42:42<+discordbot> and btw 'Skip animation' text should be +1px up 20180407 14:42:52<+discordbot> yeah, i dunno why it's not 20180407 14:48:19< grzywacz> These big cities beg for multihex, impassable buildings. ;) 20180407 14:48:42<+discordbot> I must go AFK 😦 , see you later 20180407 15:27:41< celticminstrel> @Vultraz — Would you prefer if I cherry-picked the new features from the schema branch into master before merging it> 20180407 15:27:42< celticminstrel> ^? 20180407 15:27:53< celticminstrel> I mean new features in the engine, not new features in the schema system. 20180407 16:14:07-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev 20180407 16:16:18-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.] 20180407 16:26:00< irker393> wesnoth: Celtic Minstrel wesnoth:master c62a4820fcab / src/game_events/action_wml.cpp: Deprecate [replace_map]map= in favour of [replace_map]map_data= https://github.com/wesnoth/wesnoth/commit/c62a4820fcab36e3c556f698d60e1afa9d242f84 20180407 16:26:02< irker393> wesnoth: Celtic Minstrel wesnoth:master 46f4b514822a / src/units/attack_type.cpp: SWF: Support lists for special= and special_active= https://github.com/wesnoth/wesnoth/commit/46f4b514822aadfed173f705ec401439b6aae74d 20180407 16:26:04< irker393> wesnoth: Celtic Minstrel wesnoth:master 62d39d653230 / src/units/filter.cpp: Support usage= in SUF https://github.com/wesnoth/wesnoth/commit/62d39d653230f98100a5de5aac9e8c7be2b7e7ea 20180407 16:26:06< irker393> wesnoth: Celtic Minstrel wesnoth:master 4b2dd68595a5 / src/units/filter.cpp: Support alignment= in SUF https://github.com/wesnoth/wesnoth/commit/4b2dd68595a55b5fea68a26f28a464b312a82cf4 20180407 16:26:08< irker393> wesnoth: Celtic Minstrel wesnoth:master 8581637b5146 / changelog.md: Update changelog https://github.com/wesnoth/wesnoth/commit/8581637b5146152fdd14af9ac4372a63995195e7 20180407 16:26:10< celticminstrel> Since you didn't answer, I just went ahead and did so. 20180407 16:35:05< celticminstrel> zookeeper, @Pentarctagon, please take a look at 2793. 20180407 16:35:13< celticminstrel> I'd pink Sigurd too if I knew how. 20180407 16:37:01< celticminstrel> ^ping 20180407 16:37:51<+discordbot> anything in particular, or I should just start working my way through all 49 commits? 20180407 16:37:59< celticminstrel> Also, if I could get someone to mute irker again, I could push the last couple of things to that PR. 20180407 16:38:36< celticminstrel> @Pentarctagon - Probably just work your way through. 20180407 16:38:43<+discordbot> ok 20180407 16:38:47< celticminstrel> I can't think of anything specific that I'd need you specifically to look at. 20180407 16:38:58< celticminstrel> (The last couple of commits in question are just test scenarios FTR.) 20180407 16:39:08< celticminstrel> (I mean the ones I haven't pushed yet.) 20180407 16:40:01<+discordbot> heh, me, sigurd, and zookeeper are all listed twice for the review requests. 20180407 16:40:05<+discordbot> weird 20180407 16:40:11< celticminstrel> Yeah I dunno how that happened. 20180407 16:50:20-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection] 20180407 16:51:57-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev 20180407 16:53:59-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!] 20180407 16:54:06-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection] 20180407 16:58:00-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev 20180407 17:07:22< irker393> wesnoth/wesnoth:opengl Jyrki Vesterinen 3e96a3aa2c Add a shader class AppVeyor: All builds passed 20180407 17:07:30< celticminstrel> Oh. 20180407 17:10:07< celticminstrel> @Jyrkive Any thoughts on adding functions to set parameter values in the shader? 20180407 17:10:28< celticminstrel> eg Like in https://github.com/CelticMinstrel/wesnoth-ipf/blob/master/shader.hpp 20180407 17:10:36<+discordbot> We need some way to do that, yes. 20180407 17:10:58<+discordbot> I haven't yet thought what the interface would look like or how it would be implemented. 20180407 17:11:13< celticminstrel> Okay. 20180407 17:12:37<+discordbot> Your implementation actually looks quite useful to me. Thanks. 😃 20180407 17:13:15< celticminstrel> :) 20180407 17:13:16<+discordbot> (But I don't plan to work on uniform support for a long time. Getting stuff to draw has higher priority, and the default shader won't need uniforms.) 20180407 17:14:00< celticminstrel> Uniform is official term for parameters? 20180407 17:14:03< celticminstrel> Or is that attrib? 20180407 17:14:18<+discordbot> AFAIK, uniform is the official term. 20180407 17:14:35<+discordbot> (Then there are also vertex parameters, but I don't plan to use them.) 20180407 17:14:47< celticminstrel> Vertex parameters? 20180407 17:15:22< celticminstrel> Pretty sure the ones we'll need for the IPF->shader converter are the ones that are available to both vertex and fragment shader. 20180407 17:15:54<+discordbot> OpenGL allows each vertex to carry as much data as we want. The vertex layout isn't hardcoded. I refer to anything other than vertex position or texture position as "vertex parameters". 20180407 17:16:09< celticminstrel> Interesting. 20180407 17:16:46< celticminstrel> It used to be hard-coded, didn't it? 20180407 17:16:51< celticminstrel> Like in 1.2 or whatever. 20180407 17:17:09<+discordbot> Yes, I think the vertex layout was hardcoded in OpenGL 1.x days. 20180407 17:17:24< celticminstrel> With color and one or two other things in addition to position and texpos. 20180407 17:18:02<+discordbot> Yep. IMO, vertex coloring is pretty much useless together with textures, and thus the vertex layout I defined omits it. 20180407 17:18:27<+discordbot> (We can obviously implement a coloring shader, we just need to pass color data as a uniform.) 20180407 17:19:15<+discordbot> @hrubymar10 I thought that PR was about xcode 20180407 17:19:32< celticminstrel> I think I'm doing exactly that for the IPF converter. 20180407 17:21:35< celticminstrel> ...oh, you can actually make OpenGL use only 2D coordinates in the vertex? 20180407 17:22:12<+discordbot> It doesn't really matter as long as we can pass X and Y coordinates. 20180407 17:22:50<+discordbot> The vertex shader can specify how many coordinates the position is supposed to have, and even a mismatch between the vertex and shader's expectation is allowed. 20180407 17:23:16<+discordbot> If the shader expects more coordinates than the vertex contains, Y, Z and W become 0, 0 and 1, respectively. 20180407 17:23:57-!- octalot [~steve@178.115.128.204.wireless.dyn.drei.com] has quit [Ping timeout: 240 seconds] 20180407 17:24:38< celticminstrel> I think my shaders were assuming ... 2D coordinates probably? I can't remember. 20180407 17:25:13-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev 20180407 17:25:48< celticminstrel> Looks like 2D indeed. 20180407 17:35:00-!- octalot [~steve@178.115.128.204.wireless.dyn.drei.com] has joined #wesnoth-dev 20180407 18:43:10< irker393> wesnoth/wesnoth:1.14 loonycyborg 823d37c37a nsi script: update expected extensions o AppVeyor: All builds passed 20180407 18:48:26< celticminstrel> @Vultraz — What do you prefer for overridden functions? 20180407 18:48:33< celticminstrel> "virtual void f() override" 20180407 18:48:38< celticminstrel> "void f() override" 20180407 19:15:30< vn971> I wonder about the situation behind the wesnoth.remove_modification method. It was undocumented up until today. I found it by checking out wml/object.lua code, and documented it: https://wiki.wesnoth.org/LuaWML/Units#wesnoth.remove_modification 20180407 19:15:30< vn971> However, there is a problem. I see now that the implementation is very anti-Lua-ish. (To be honest, I do not understand why would anyone use it at all instead of [remove_object].) It can me made much nicer though if we make the second argument of wesnoth.remove_modification be a string, instead of {id=string}. 20180407 19:15:30< vn971> So, should we try to do that now? Change the syntax of the Lua method wesnoth.remove_modification? 20180407 19:16:05< celticminstrel> The second argument is a filter/ 20180407 19:16:19< celticminstrel> You could make it also support a string as a shorthand for id=string, I suppose. 20180407 19:16:26< celticminstrel> Or two strings as a shorthand for string1=string2. 20180407 19:16:58< celticminstrel> But the filter in remove_modification supports the full power of [filter_wml], so I don't think we should take that away just because the common use-case doesn't need that power. 20180407 19:17:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 245 seconds] 20180407 19:18:24< vn971> celticminstrel: wait, so the second arguments can filter objects by other stuff, not only the object-s id ? 20180407 19:18:30< celticminstrel> Yes. 20180407 19:18:33< celticminstrel> Editing wiki now. 20180407 19:18:53< vn971> celticminstrel: got it, thanks. Yes, if you mean my part I created today, it should be edited (I was about to fix it myself). 20180407 19:19:39< vn971> celticminstrel: I was confused by [remove_object] which only accepts object id: https://wiki.wesnoth.org/DirectActionsWML#.5Bremove_object.5D 20180407 19:20:12< celticminstrel> Yeah that's why I announced my intention, because I figured you would also want to edit it. If you want you can still add a link to somewhere that documents [filter_wml], though there's no good place for that ATM (perhaps it should be moved from StandardUnitFilter to FilterWML?). 20180407 19:22:20-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!] 20180407 19:22:24< celticminstrel> For example, you could remove all modifications that affect a unit's hitpoints with unit:remove_modification{wml.tag.effect{apply_to='hitpoints'}} 20180407 19:22:40< celticminstrel> (Note that that would even remove the resilient trait, for example.) 20180407 19:30:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180407 20:03:50-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev 20180407 21:15:25< celticminstrel> I have now replaced all [key] definitions in the GU2 schema with the DEFAULT_KEY or REQUIRED_KEY macro. 20180407 21:15:48< celticminstrel> (GUI2 doesn't have any optional keys without a specified default.) 20180407 21:18:51< celticminstrel> Working on validating the schema itself now. 20180407 21:35:48-!- gallaecio [~quassel@57.99.79.188.dynamic.jazztel.es] has quit [Remote host closed the connection] 20180407 21:43:34-!- irker393 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout] 20180407 21:54:56< celticminstrel> So does @Pentarctagon's silence on the PR mean there was nothing worth commenting on, or have you not gotten to it yet? 20180407 21:55:51<+discordbot> I looked through it, though I'm not really sure why I'm a reviewer to be honest. 20180407 21:56:44< celticminstrel> Well, that's good enough, I guess. 20180407 22:32:56-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 22:33:02-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 22:36:35-!- grzywacz [~karol@wesnoth/developer/grzywacz] has quit [Ping timeout: 240 seconds] 20180407 22:39:13<+discordbot> Celmin: former 20180407 22:39:37<+discordbot> With virtual keyword 20180407 22:39:50< celticminstrel> I think that's what I went with, so that's good. 20180407 22:40:06<+discordbot> Makes it clearer 20180407 23:02:14-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180407 23:02:21-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180407 23:07:39< octalot> I'm wondering how long to wait for an addon to be uploaded to 1.14, before just hijacking it and uploading it myself. Specifically Valley of the Ancients (Dimaga Beach). And no, I shouldn't have enough time to consider maintaining it... 20180407 23:08:17< celticminstrel> Well, I'd say at least wait until after the 1.14 release. 20180407 23:09:16-!- irker741 [~irker@uruz.ai0867.net] has joined #wesnoth-dev 20180407 23:09:16< irker741> wesnoth/wesnoth:master newfrenchy83 0772e03dd3 Update wesnoth.cbp AppVeyor: All builds passed 20180407 23:09:28< octalot> sounds fair 20180407 23:16:22<+discordbot> Celmin: honestly I still get confused at the difference between a virtual override and a function that shadows one in the base class. So I find having the virtual keyword there to be helpful 20180407 23:22:40< celticminstrel> I think a large part of the debug build being slow might be simply having the debugger attached... 20180407 23:25:45< celticminstrel> Hm, tutorial test is inconclusive (and probably not a good example anyway), but something you could try @Vultraz is using a debug build but runing without the debugger. You'd still be able to jump into the debugger if you hit an error or something. 20180407 23:25:54< celticminstrel> (This is assuming you're using MSVC.) 20180407 23:26:15<+discordbot> What? 20180407 23:26:24< celticminstrel> The default keystroke for that is Ctrl+F5, it's an option in the Debug menu. 20180407 23:26:31< celticminstrel> "Start without Debugging" 20180407 23:26:46< celticminstrel> You mentioned testing on release builds because debug is slow, did you not? 20180407 23:26:53< celticminstrel> Maybe my information is old. 20180407 23:27:13< celticminstrel> Still, "Start without Debugging" is a useful thing to know about anyway. 20180407 23:29:21<+discordbot> What are you asking me to do 20180407 23:29:24<+discordbot> And why 20180407 23:30:06< celticminstrel> I'm pointing out that you can run a debug build without the debugger, which might help with the slowness. Or it might not. I'm not sure. I'm just suggesting you try it. 20180407 23:30:21<+discordbot> I know you can do it 20180407 23:30:25<+discordbot> I’ve done it before 20180407 23:30:34<+discordbot> But why would I need to do it now? 20180407 23:30:51< celticminstrel> Okay, if you already know, then never mind! 20180407 23:31:03<+discordbot> Did something break? 20180407 23:31:13< celticminstrel> No, not that I know of. 20180407 23:31:13-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection] 20180407 23:31:26< celticminstrel> I was just trying to be helpful. :/ 20180407 23:31:49<+discordbot> Oh 20180407 23:31:54<+discordbot> Ok then. Thanks 20180407 23:32:05<+discordbot> I thought you were asking me to debug something in the tutorial 20180407 23:35:57-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds] 20180407 23:36:16<+discordbot> Celmin: the main issue is I launch wesnoth via shortcut with certain cl options like wconsole pre set 20180407 23:36:36<+discordbot> And to use SWD you need to launch from within VS 20180407 23:38:12< celticminstrel> Wait what? 20180407 23:38:17< celticminstrel> Where do you launch from? 20180407 23:38:34<+discordbot> Taskbar 20180407 23:39:00< celticminstrel> So that's basically the same as SWD, right? The debugger shouldn't be attached in that case. 20180407 23:45:50-!- gfgtdf [~chatzilla@x4e36316a.dyn.telefonica.de] has joined #wesnoth-dev 20180407 23:46:21< gfgtdf> you can also configure the commandlien options that msvc passes to wesntoh.exe when starting it from visualstudio 20180407 23:56:26-!- vn9711 [~vasya@94.158.103.15] has joined #wesnoth-dev 20180407 23:59:35-!- vn971 [~vasya@94.158.103.15] has quit [Ping timeout: 265 seconds] --- Log closed Sun Apr 08 00:00:35 2018