--- Log opened Tue May 22 00:00:31 2018 20180522 00:01:53<+discordbot> There appears to be a very subtle change in the WML event handling sequence in 1.14.x. 20180522 00:03:31< celticminstrel> I'm absolutely unsurprised. 20180522 00:03:32<+discordbot> Oh wait, never mind that. 20180522 00:03:43<+discordbot> It's the same in 1.12, I just never realized. 20180522 00:03:48< celticminstrel> Oh. 20180522 00:05:21< irker078> wesnoth: ln-zookeeper wesnoth:1.14 bed524a0a3b7 / data/campaigns/The_South_Guard/ (maps/03_A_Desperate_Errand.map scenarios/03_A_Desparate_Errand.cfg): TSG S03: Made the scenario very subtly harder https://github.com/wesnoth/wesnoth/commit/bed524a0a3b7511db700dc1ef1d59c6a461bc374 20180522 00:05:23<+discordbot> When handling events involving unit movement, the unit's bars aren't updated on the screen. 20180522 00:05:44<+discordbot> So the unit displays bar-less. But it turns out this was already the case in 1.12, so it's not a regression. 20180522 00:11:34< irker078> wesnoth: ln-zookeeper wesnoth:1.14 f320a0ee5d57 / data/campaigns/Heir_To_The_Throne/scenarios/08_The_Princess_of_Wesnoth.cfg: HttT S08: Increased Li'sar's income slightly https://github.com/wesnoth/wesnoth/commit/f320a0ee5d57d22f8efcc6b93cbcdf56846e30c9 20180522 00:14:04<+discordbot> Actually, that's for sighted events. 20180522 00:14:13<+discordbot> moveto in 1.12 has the bars drawn. 20180522 00:14:48<+discordbot> While it doesn't in 1.14. 20180522 00:15:57< irker078> wesnoth: ln-zookeeper wesnoth:1.14 c72455e9b0ed / data/campaigns/The_Rise_Of_Wesnoth/scenarios/15_A_New_Land.cfg: TRoW S15: Increased the amount of lvl2 units enemies recruit https://github.com/wesnoth/wesnoth/commit/c72455e9b0ed50a242f2e897092ff2528f166ee4 20180522 00:18:35< zookeeper> so are you now pushing regular translation updates through steam, and if so, are some times better/worse for committing small string changes? 20180522 00:20:08< celticminstrel> Probably. 20180522 00:20:59< celticminstrel> IMO translation updates between main releases should push only the translations folder, no actual string changes, but I dunno if that's what he's doing. 20180522 00:22:23<+discordbot> That doesn't work to well currently though because pot-updates don't fit into the process 20180522 00:23:46<+discordbot> It a question of either making no pot updates, which keeps translations cimpatible to the steam version, or make them which allows translators to translate the newer changes 20180522 00:24:00<+discordbot> You mean what the packagers are doing, but yes, they are only updating the translations folder. 20180522 00:27:45<+discordbot> Seems having the bars not drawn in moveto events is an acceptable change 20180522 00:28:03<+discordbot> There are no pot updates in stable except at the start of the string freeze and right before a release, so it'd just be a matter of asking packagers to not rebuild translations during the string freeze. 20180522 00:28:56< zookeeper> well, i guess i'm done, i don't even think those last balance tweaks are important enough to changelog... 20180522 00:30:04<+discordbot> As far as I understood pot updates are involved when Ivanovic merges translation updates 20180522 00:30:17<+discordbot> No, they are not. 20180522 00:30:45<+discordbot> po updates are, which is to say, the pot is merged into the po files. 20180522 00:34:24<+discordbot> Yes, I know, though there is nothing which stops you from recreating the pot first. If that happens or not - I'm not entirely sure, but my impression was that this is way things are handled currently. 20180522 00:35:46< zookeeper> @Vultraz, feel free to merge the UtBS PR's at your own peril, and you should totally fix the shyde TC 20180522 00:36:52< irker078> wesnoth: ln-zookeeper wesnoth:1.14 4941c4b461cb / changelog.md players_changelog.md: Updated changelogs https://github.com/wesnoth/wesnoth/commit/4941c4b461cbce5f824cea649b75fab4a0fe4d62 20180522 00:38:15<+discordbot> @Vultraz I’ll try to submit my PR tonight for hotkeys 20180522 00:38:35<+discordbot> Shyde tc? 20180522 00:39:01< zookeeper> quen 20180522 00:39:09<+discordbot> What? 20180522 00:39:12< zookeeper> @Vultraz, uh, the Quenoth Shyde TC is broken... 20180522 00:39:21<+discordbot> Uh? 20180522 00:40:00< zookeeper> you didn't commit a sprite without looking at it in-game, did you? :p 20180522 00:40:41<+discordbot> yes? 20180522 00:41:31<+discordbot> HUH 20180522 00:41:36<+discordbot> that is indeed broken 20180522 00:41:37< zookeeper> anyways, off to bed -> 20180522 00:41:38<+discordbot> odd 20180522 00:41:40< irker078> wesnoth: Severin Glöckner wesnoth:1.14 9568d48ff77d / changelog.md players_changelog.md: Updated Changelogs https://github.com/wesnoth/wesnoth/commit/9568d48ff77da8c3f5496a0559845419558b1d11 20180522 00:42:13<+discordbot> u slippin 20180522 00:42:13< zookeeper> (unless you really need me to do something else real quick) 20180522 00:42:41<+discordbot> nah 20180522 00:42:53<+discordbot> anyway, no release today since ancestral's hotkey changes aren't in 20180522 00:45:10<+discordbot> @sevu Only loonycyborg and I are in charge of pot updates right now. 20180522 00:45:31<+discordbot> @shadowm option the second 20180522 00:45:31<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448285242231357454/unknown.png 20180522 00:45:35<+discordbot> To my knowledge no-one does a pot update when merging translation updates, and if they do they shouldn't. 20180522 00:45:45<+discordbot> @Vultraz Ew, no. 20180522 00:45:53<+discordbot> I hate light fonts with a passion. 20180522 00:46:09<+discordbot> 😦 20180522 00:46:24<+discordbot> You already knew this for years. 20180522 00:46:30-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 245 seconds] 20180522 00:46:42<+discordbot> meh 20180522 00:46:44<+discordbot> no changes, then 20180522 00:47:13<+discordbot> or maybe I can just make the size larger 20180522 00:49:20<+discordbot> option the third 20180522 00:49:21<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448286205373579285/unknown.png 20180522 00:49:50<+discordbot> much better 20180522 00:50:03< celticminstrel> What was option the first? 20180522 00:50:21< celticminstrel> This? https://cdn.discordapp.com/attachments/259976436490829825/448257585280319488/unknown.png 20180522 00:50:27<+discordbot> Also, you realize that I have set a hard size on that text? 20180522 00:50:39< irker078> wesnoth/wesnoth:master Severin Glöckner f291deb6cf LoW 2 S7 - fix repeated endlevel executi AppVeyor: All builds passed 20180522 00:51:10<+discordbot> yes 20180522 00:51:12<+discordbot> https://github.com/project-ethea/Naia/blob/master/macros/storytxt-common.cfg#L39 20180522 00:51:17<+discordbot> to both of you 20180522 00:51:18<+discordbot> Size 24. 20180522 00:51:31<+discordbot> (Since title labels are size 22.) 20180522 00:51:42< celticminstrel> Then I like the first option best. 20180522 00:51:44<+discordbot> this is 36 20180522 00:51:49< celticminstrel> The other two are plain and boring. 20180522 00:51:54<+discordbot> Oh for the love of. 20180522 00:52:49<+discordbot> 36 is way too large. 20180522 00:53:08<+discordbot> Also as a reminder, your screenshots are useless to me because they are not at my native pixel density. 20180522 00:53:30<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448287251865272320/unknown.png 20180522 00:53:31<+discordbot> it's not way too large 20180522 00:53:37<+discordbot> Yes it is. 20180522 00:53:46<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448287318089138189/unknown.png 20180522 00:54:01<+discordbot> The main text is tiny in comparison. 20180522 00:54:47<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448287574327558165/unknown.png 20180522 00:55:20<+discordbot> This is size 24: 20180522 00:55:20<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448287712328417284/unknown.png 20180522 01:05:41-!- travis-ci [~travis-ci@ec2-54-158-234-7.compute-1.amazonaws.com] has joined #wesnoth-dev 20180522 01:05:42< travis-ci> wesnoth/wesnoth#18299 (1.14 - c72455e : ln-zookeeper): The build passed. 20180522 01:05:42< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/381929501 20180522 01:05:42-!- travis-ci [~travis-ci@ec2-54-158-234-7.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180522 01:11:38<+discordbot> @shadowm btw, did you make any progress on the Eloh sprite? 20180522 01:11:51<+discordbot> Uhhhhhh. 20180522 01:11:58<+discordbot> I completely forgot about it. 20180522 01:13:18<+discordbot> In my defense, two situations that you are already aware of (one being the 1.14 release) as well as porting IftU/AtS/Naia to 1.14 completely overshadowed everything else. Is there anything besides this that I've forgotten? 20180522 01:15:51<+discordbot> (And I'm not even half-done with the last one of the above.) 20180522 01:16:21-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20180522 01:17:25<+discordbot> No, I think that’s the only thing 20180522 01:17:53<+discordbot> I'm pretty sure I'm also supposed to draw stuff for Yumi. 20180522 01:18:13<+discordbot> I remembered that this morning. 20180522 01:19:05<+discordbot> I literally woke up with the thought "wait wasn't I supposed to draw a few Guardians for Yumi? which one is the first in the queue, was it Valdir?" 20180522 01:21:57<+discordbot> does he wear a suit 20180522 01:22:27<+discordbot> I'm not sure what kind of question is that. 20180522 01:22:39<+discordbot> Oh I think I remember now. 20180522 01:22:58<+discordbot> You told me five years ago that Ethea was basically Earth 2050. 20180522 01:23:03<+discordbot> You're thinking of the Guardian of Water of the Second Cycle. 20180522 01:23:09<+discordbot> so logically, its guardian would wear a suit 20180522 01:23:15<+discordbot> Yumi is working with First Cycle story material. 20180522 01:23:47-!- travis-ci [~travis-ci@ec2-54-227-75-141.compute-1.amazonaws.com] has joined #wesnoth-dev 20180522 01:23:48< travis-ci> wesnoth/wesnoth#18301 (1.14 - 9568d48 : Severin Glöckner): The build passed. 20180522 01:23:48< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/381935872 20180522 01:23:48-!- travis-ci [~travis-ci@ec2-54-227-75-141.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180522 01:24:21<+discordbot> Hopefully that doesn't mean he'll be dressed like an 18th century French aristocrat 20180522 01:24:50<+discordbot> He's not really a human-type species so no. 20180522 01:25:38<+discordbot> Oh dear. 20180522 01:25:55<+discordbot> I suddenly found on my game screen a skeleton bobbing its butt towards the camera. 20180522 01:26:11<+discordbot> Turns out it's the new north-facing Revenant standing animation. 20180522 01:26:22<+discordbot> It has a slight TC malfunction in one frame. 20180522 01:26:56<+discordbot> twerking skeletons :thonk: 20180522 01:29:25<+discordbot> Valdir first is preferable, but don't worry about it for now 20180522 01:29:53<+discordbot> Genesis II is low on my work queue and will get pushed back even more if the board decides to have me redo UtBS or some other campaign 20180522 01:30:23<+discordbot> I need to go read the THoT thread 20180522 01:30:37<+discordbot> MEME 👏 REVIEW 👏 20180522 01:30:41< irker078> wesnoth: Iris Morelle wesnoth:1.14 bfeb22c24cac / changelog.md data/core/images/units/undead-skeletal/revenant/revenant-ne-bob3.png: Fix broken TC in a north-facing standing animation frame for the Revenant https://github.com/wesnoth/wesnoth/commit/bfeb22c24cac9eb14e6c13638abe1c547ecdad17 20180522 01:30:43<+discordbot> heh 20180522 01:31:08<+discordbot> I vote we change Vultraz's project member role to "project memer" 20180522 01:31:18<+discordbot> twitches 20180522 01:31:35<+discordbot> 👌 20180522 01:32:16< irker078> wesnoth: Iris Morelle wesnoth:master df782f5401f2 / changelog.md data/core/images/units/undead-skeletal/revenant/revenant-ne-bob3.png: Fix broken TC in a north-facing standing animation frame for the Revenant https://github.com/wesnoth/wesnoth/commit/df782f5401f2f1fc8b6b80d30a03604ff4a1caad 20180522 01:33:07<+discordbot> hi, excuse me, I just found Wesonth on steam and still developing so I started crying I'm seriously happy, I played it for the first time around 10 years ago, even made some friends I still talk to by playing it this game is always installed on my phones and pc c: just wanted to say thanks a lot 20180522 01:33:25<+discordbot> 😃 20180522 01:33:53<+discordbot> can I meet the original developers here? o,o 20180522 01:34:59<+discordbot> Dave is still around 20180522 01:35:19<+discordbot> None of us are the original devs, but some of us have been here for a long time 20180522 01:36:00<+discordbot> I still love you ❤ 20180522 01:36:53<+discordbot> but who is responsible for all the changes i hate? 20180522 01:37:14<+discordbot> depends on what changes those might be 😛 20180522 01:39:03<+discordbot> idk anymore 20180522 01:39:17<+discordbot> you guys are way better than dungeon crawl stone soup devs so y'all good 20180522 01:41:04<+discordbot> They release more often than us though 20180522 01:41:07<+discordbot> yes 20180522 01:41:23<+discordbot> i'd prefer if they kept their trash releases to themselves though 20180522 01:44:15<+discordbot> btw what do you focus on developing now? can I find patch notes or something like that? 20180522 01:46:02<+discordbot> Massive refactor to the rendering backend 20180522 01:50:47< celticminstrel> Oh yeah @Vultraz, are we still using [+units] instead of [units] in the campaigns? 20180522 01:50:56<+discordbot> don't know 20180522 01:51:05< celticminstrel> Something to look at when you have a moment. 20180522 01:51:45<+discordbot> don't know how to grep for it 20180522 01:51:59< celticminstrel> ... 20180522 01:52:14< celticminstrel> git grep -F "[+units]" 20180522 01:52:32<+discordbot> fgrep '[+units]'data/campaigns/*/_main.cfg` 20180522 01:53:18< celticminstrel> Space seems to be missing before "data" 20180522 01:53:22<+discordbot> nothing 20180522 01:53:26<+discordbot> Fixed it. 20180522 01:53:26< celticminstrel> Good good 20180522 01:54:00<+discordbot> when I grep for something I almost always use grep foo -ri * 20180522 01:54:07<+discordbot> doesn't always give me what I want, though 20180522 01:54:19<+discordbot> fgrep is an alias for grep -F. 20180522 01:54:24< celticminstrel> Well, I added -F to make it not interpret the search string as a regex. 20180522 01:54:30<+discordbot> ohhh 20180522 01:54:34<+discordbot> no wonder that was happening... 20180522 01:54:37< celticminstrel> So the [] are matched literally rather than defining a character class. 20180522 01:54:51<+discordbot> Otherwise I use egrep instead, which is an alias for grep -E. 20180522 01:55:01<+discordbot> Extended regexes supported by grep -E are far easier to type. 20180522 01:55:25< celticminstrel> I usually use -E but sometimes need -P instead. 20180522 02:04:13<+discordbot> I occurs to me that since we have [return] now, we can start writing WML that exits events early based on negative conditions... 20180522 02:04:20<+discordbot> instead of deep indent with [if] 20180522 02:06:32<+discordbot> it also occurs to me there's nothing actually stopping us from writing all events in Lua 🤔 20180522 02:06:36<+discordbot> [lua] is actionwml 20180522 02:06:57<+discordbot> Somehow I managed to realize that 5 years ago. 20180522 02:08:24<+discordbot> I wonder why we don't do it 20180522 02:08:40<+discordbot> Because existing code uses WML. 20180522 02:09:10<+discordbot> WML is also the norm for Wesnoth content creators and mainline devs. 20180522 02:09:37<+discordbot> (And no I don't think this is the right time to have a rehash of the whole "let's replace wml" debate.) 20180522 02:10:36<+discordbot> No, but like 20180522 02:10:37<+discordbot> in UMC 20180522 02:10:43<+discordbot> YOU could have done it 20180522 02:10:49<+discordbot> I could have done it 20180522 02:11:35-!- sevu [~Shiki@p54854120.dip0.t-ipconnect.de] has quit [Remote host closed the connection] 20180522 02:12:58<+discordbot> I can't, I have the mental capacity of a chicken 20180522 02:13:01<+discordbot> I guess Lua is a bit unwieldy when tables need to be passed arond 20180522 02:13:13<+discordbot> The Lua API was still a moving target when I finished AtS E1. 20180522 02:14:34<+discordbot> Only after that it started to become more of an official thing, but by then I had an established framework (which has since become Naia) and rewriting code would have required me to give up what was at the time an unprecedented bout of motivation to work on AtS. 20180522 02:15:18<+discordbot> nods 20180522 02:15:27<+discordbot> WML is also easier than lua (at least to me) 20180522 02:15:44<+discordbot> I hate having to write closing tags 20180522 02:15:49<+discordbot> don't need to discourage content creation with harder languages 20180522 02:16:07<+discordbot> The only aspect of Lua that is "hard" is the 1-based list subscripts. 20180522 02:16:22<+discordbot> at least 10% of the reason I want to switch to a bracket-based syntax is to eliminate the need for typing closing tags. 20180522 02:16:23<+discordbot> The rest is pretty simple compared to even Javascript. 20180522 02:16:57<+discordbot> I don't really care much about the language. 20180522 02:17:33<+discordbot> In my brain I work with what the code says, not how it's written. 20180522 02:18:01<+discordbot> Certainly 20180522 02:18:14<+discordbot> That's why I hate it when people ask me programming questions that focus on the syntax or the language employed and not the code itself. 20180522 02:18:29<+discordbot> looks in vultraz's general direction 20180522 02:18:41<+discordbot> But that becomes harder when you manually need to go type the closing tag because your editor doesn't autocomplete it. 20180522 02:18:50<+discordbot> It constantly takes me out of the logic 20180522 02:18:54<+discordbot> I'm more concerned about the API than the language. 20180522 02:19:20<+discordbot> Like I was saying when I completed AtS E1 our support for Lua was still a moving target and the API was definitely not as rich as it is now. 20180522 02:19:32<+discordbot> And things like maintainability weren't a concern back then. 20180522 02:19:44<+discordbot> thank calmin and iceiceice fo rlike 20180522 02:19:52<+discordbot> 80% of the featues it now has 20180522 02:19:55<+discordbot> Shortly afterwards there was a new stable series with the improved Lua engine. 20180522 02:20:06<+discordbot> That was 1.10. 20180522 02:20:38<+discordbot> From that point onwards the Lua API snowballed because people took a liking to it (myself included) and started to contribute more features for it to the game. 20180522 02:20:46<+discordbot> (the lua gui engine is still crap and needs a redesign) 20180522 02:20:54<+discordbot> Lua GUI2 API. 20180522 02:21:03<+discordbot> That wasn't even officially-sanctioned as you know. 20180522 02:21:17<+discordbot> I know 20180522 02:21:22<+discordbot> it shows 20180522 02:21:24<+discordbot> It was implemented in a moving target platform and targeted a moving target API. 20180522 02:21:32<+discordbot> (The Lua engine and GUI2, respectively.) 20180522 02:21:51<+discordbot> By all means, exposing GUI2 to content developers was lunacy. 20180522 02:22:12<+discordbot> If GUI2's development hadn't started to stall, it would have been broken a lot. 20180522 02:22:13<+discordbot> I was the first to use it for anything more than just a simple prompt 20180522 02:22:18<+discordbot> I first saw it was possible in the WLP 20180522 02:22:36<+discordbot> and expanded on its use. But I kept finding it couldn't do a whole bunch of stuff 20180522 02:23:02<+discordbot> But since it was already slowing down because of its own design issues impeding further development, the Lua GUI2 API did not suffer from the gravest consequences of implementing a wrapper for an unstable API. 20180522 02:23:19<+discordbot> And here we are now. 20180522 02:23:27<+discordbot> Historical note: the Item Pickup prompt in SoD was based o the WLP prompt of a similar nature 20180522 02:23:34<+discordbot> GUI2 is still a mess and the Lua bindings are still unwieldy but they do the job, most of the time. 20180522 02:23:40<+discordbot> I think i copied it and added a few more things 20180522 02:24:54<+discordbot> but a more complex use of GUI2 dialogs in UMC is probably the one trend I can safely say I started. 20180522 02:25:21<+discordbot> I think gfgtdf is a contender to that title as well. 20180522 02:25:59<+discordbot> I will say that API subset was stagnating for a long while after silene's departure and I was the first to make a move to improve it. 20180522 02:26:17<+discordbot> (at the very least it's probably one of the few things in my campaigns I didn't copy from your campaigns) 20180522 02:26:22<+discordbot> (Just exposing the markup property after a previous patch from a third party was rejected by mordante himself.) 20180522 02:38:06<+discordbot> HELLO 20180522 02:38:17<+discordbot> Since when does [message] scrolling to units ignore shroud? 20180522 02:39:45<+discordbot> odd 20180522 02:39:53<+discordbot> This is not good. 20180522 02:40:11<+discordbot> It exposes the location of the speaker through shroud. 20180522 02:41:19<+discordbot> try changing data/lua/wml/message.lua:376, argument 3 to true 20180522 02:42:32<+discordbot> If you put it like that the obvious answer is that it must fix it. 20180522 02:42:56<+discordbot> (I looked it up since I'm not into voodoo fixes.) 20180522 02:43:24<+discordbot> eh? 20180522 02:44:18<+discordbot> What is the question in this case? 20180522 02:44:45<+discordbot> Also yes, it fixes the issue, obviously. 20180522 02:44:58<+discordbot> I'd prefer if you committed the fix yourself in this case though. 20180522 02:45:15<+discordbot> Also, it turns out that in both 1.12 and 1.14 this still exposes the coordinates for the speaker. 20180522 02:45:35< celticminstrel> And Vultraz misspelled my shortname, fun. 20180522 02:45:43<+discordbot> While the game doesn't scroll the view to the speaking unit, it still appears to select the hex through shroud and the coordinates display in the status area. 20180522 02:46:43< celticminstrel> Should probably have a check like "if is_shrouded(unit_location) then highlight = false end" 20180522 02:46:55< celticminstrel> That's pseudocode though, I dunno how to actually implement it. 20180522 02:48:25<+discordbot> The problem in this case lies in the implementation of wesnoth.scroll_to_tile or one of the functions it depends on internally. 20180522 02:49:15<+discordbot> But since it's not a regression it's hard to say how much of an issue it is. 20180522 02:49:31< celticminstrel> Well, I dunno, I think the highlighting of the hex is actually separate from the scrolling? 20180522 02:49:42< celticminstrel> It's handled by wesnoth.highlight_hex IIRC 20180522 02:49:49<+discordbot> Oh, right, it is now. 20180522 02:50:15< celticminstrel> But it's actually hilited even in 1.12, you said? 20180522 02:50:22<+discordbot> Yes. 20180522 02:50:44< celticminstrel> Huh. 20180522 02:52:10< celticminstrel> But yeah, adding only_if_visible to the scroll_to_tile call seems like a good idea. 20180522 02:52:25<+discordbot> It's not just a good idea, it's a must. :p 20180522 02:52:40< celticminstrel> Whatever. 20180522 02:52:58<+discordbot> Do you disagree with this behaviour change from 1.12 being unwanted? 20180522 02:53:04< celticminstrel> No. 20180522 02:53:22<+discordbot> Okay, wasn't sure how to interpret that "whatever". 20180522 03:22:54<+discordbot> how is it that SDL_Mixer has no documentation 20180522 03:23:22< irker078> wesnoth: Iris Morelle wesnoth:1.14 d2e94b320fe1 / changelog.md data/lua/wml/message.lua: Do not allow [message] to scroll to units through fog/shroud https://github.com/wesnoth/wesnoth/commit/d2e94b320fe130ef07d0fc9f7ae5967a027d8e51 20180522 03:23:29< irker078> wesnoth: Iris Morelle wesnoth:master 8ae38ac19fa4 / changelog.md data/lua/wml/message.lua: Do not allow [message] to scroll to units through fog/shroud https://github.com/wesnoth/wesnoth/commit/8ae38ac19fa48ec9c380a9bce49c17dc43be9a05 20180522 03:24:25<+discordbot> It does though. 20180522 03:24:44<+discordbot> not post-2 20180522 03:24:48<+discordbot> The documentation is in the header file. 20180522 03:26:05<+discordbot> The 1.2 docs should also still be useful since IIRC they didn't add anything to it. 20180522 03:26:29<+discordbot> It's supposed to be just an "example library", etcetera. 20180522 03:26:46<+discordbot> perhaps 20180522 03:27:00<+discordbot> Image added stuff for 2 and it didn't document it. 20180522 03:27:58<+discordbot> It's probably documented in the header. 20180522 03:28:06-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity] 20180522 03:28:19<+discordbot> that is true 20180522 03:28:37<+discordbot> Of course that's probably not the kind of documentation you expect, but you can't expect much from a glorified DIY kit. 20180522 03:40:55<+discordbot> so the UI lockup is caused by Mix_FadeOutMusic 20180522 03:41:47<+discordbot> don't know if there's anything that can be done 20180522 03:51:16< irker078> wesnoth/wesnoth:1.14 gfgtdf 92f466ec81 don't generate a savegame snapshot unle AppVeyor: All builds passed 20180522 03:55:48< irker078> wesnoth: Charles Dang wesnoth:master 8d3f54ce7c46 / changelog.md data/lua/wml-tags.lua: [remove_sound_source] now accepts a comma-separated ID list https://github.com/wesnoth/wesnoth/commit/8d3f54ce7c4647a97efb8ddb8b598d5a1af0f815 20180522 03:56:29<+discordbot> I wonder if I can backport that. It would have 0 effect on existing usecases... 🤔 20180522 03:59:25<+discordbot> One thing that could be done would be to just not crossfade. 20180522 03:59:36<+discordbot> Why do I keep calling it that? It's not crossfade. 20180522 03:59:51<+discordbot> It's a fade-in followed by a fade-out. A crossfade requires two overlapping streams. 20180522 04:00:05<+discordbot> ... a fade-out followed by a fade-in rather. 20180522 04:13:13-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] 20180522 04:14:48< irker078> wesnoth: gfgtdf wesnoth:master 7eec132d762d / src/playsingle_controller.cpp: don't generate a savegame snapshot unless needed https://github.com/wesnoth/wesnoth/commit/7eec132d762dec707af0d9ea2cf49e2e1678f56c 20180522 04:14:49< irker078> wesnoth: ln-zookeeper wesnoth:master c0583a184292 / data/campaigns/Sceptre_of_Fire/ (11 files in 2 dirs): SoF S9: Fixed some issues with the volcano eruption (fixes #3099) https://github.com/wesnoth/wesnoth/commit/c0583a18429204c047aec16dbc553a1ad5cb13e0 20180522 04:14:51< irker078> wesnoth: ln-zookeeper wesnoth:master 856b8ce78b19 / data/campaigns/The_South_Guard/ (maps/03_A_Desperate_Errand.map scenarios/03_A_Desparate_Errand.cfg): TSG S03: Made the scenario very subtly harder https://github.com/wesnoth/wesnoth/commit/856b8ce78b196f1d25eeb8d5c6cc28c1b5d2597e 20180522 04:14:52<+discordbot> almost missed forward-porting that gfgtdf commit 20180522 04:14:53< irker078> wesnoth: ln-zookeeper wesnoth:master bd23bb32535c / data/campaigns/Heir_To_The_Throne/scenarios/08_The_Princess_of_Wesnoth.cfg: HttT S08: Increased Li'sar's income slightly https://github.com/wesnoth/wesnoth/commit/bd23bb32535cf69c0507a15e48b7ade5cd8b156a 20180522 04:14:55< irker078> wesnoth: ln-zookeeper wesnoth:master 0de4a4c7068c / data/campaigns/The_Rise_Of_Wesnoth/scenarios/15_A_New_Land.cfg: TRoW S15: Increased the amount of lvl2 units enemies recruit https://github.com/wesnoth/wesnoth/commit/0de4a4c7068ca82bcdae7242bd6751d03ce1fd5b 20180522 04:14:57< irker078> wesnoth: ln-zookeeper wesnoth:master 59b69288bafd / changelog.md players_changelog.md: Updated changelogs https://github.com/wesnoth/wesnoth/commit/59b69288bafd026c01dab7544d837f41ed1ee657 20180522 04:18:14<+discordbot> That's why people should forward/back-port their own stuff or file issues if they can't. 20180522 04:21:29<+discordbot> I've given up 20180522 04:21:35<+discordbot> and decided to do all porting myself 20180522 04:21:53<+discordbot> That's not fair. 20180522 04:22:14<+discordbot> (in case someone doesn't do it properly in the first place of course, like yourself) 20180522 04:23:04<+discordbot> You're spending time on a task that is supposed to be the responsibility of each individual with push access, time you could be spending coding or something something otherwise more productive. 20180522 04:23:10<+discordbot> @zookeeper there was a conflict i had to resolve in the SoF commit about a mask filter. on 1.14 you replaced Rr with Q...somethig something, but on master the filter was Rr,Dd. 20180522 04:23:14<+discordbot> I kept the Dd 20180522 04:23:17<+discordbot> is that correct 20180522 04:23:20<+discordbot> Furthermore, there's an increased risk that you forget a change or port a change that shouldn't be ported. 20180522 04:23:57<+discordbot> Yes, but the risk is far lower than just waiting for them to do it themselves 20180522 04:24:00<+discordbot> something that may never happen 20180522 04:36:55<+discordbot> So you no longer need me to investigate why switching the playlist blocks the main thread? 20180522 04:37:27<+discordbot> Judging from Vultraz's analysis, it sounds like disabling fade-out is indeed the only way to get rid of the problem. 20180522 04:37:37<+discordbot> @shadowm incidently, this is another reason I'm in favor of the rolling release model. Commits would only need to be on 1 branch. 20180522 04:37:52<+discordbot> @jyrkive not necessarily. Music fadeout is possible via WML. 20180522 04:38:09<+discordbot> SDL_Mixer just does a volume fadeout, nothing special 20180522 04:38:10<+discordbot> Rolling release model won't help you with large refactoring, unfortunately. 20180522 04:38:11<+discordbot> It's a blocking operation in WML. 20180522 04:38:36<+discordbot> fml 20180522 04:38:40<+discordbot> It's just that it blocks for a tiny amount of time at a time. 20180522 04:38:40<+discordbot> we could thread it 20180522 04:38:57<+discordbot> (I assume you're talking of my [fade_out_music] tag.) 20180522 04:39:00<+discordbot> yes 20180522 04:39:04<+discordbot> Is SDL_Mixer API documented to be thread-safe? 20180522 04:39:19<+discordbot> who knows. they don't have such detailed documentation 20180522 04:39:21<+discordbot> If not, we should assume it isn't, and we can only safely call its functions in the main thread. 20180522 04:39:28<+discordbot> I think i saw somrthere no? 20180522 04:39:30<+discordbot> somewhere* 20180522 04:40:04<+discordbot> what about OpenAL? 20180522 04:40:19<+discordbot> That would require a whole rewrite of the sound code. 20180522 04:40:23<+discordbot> (I explained this before.) 20180522 04:41:22<+discordbot> @Vultraz Indeed, rewriting sound code without SDL_Mixer would allow us to implement fade-out without blocking the main thread. 20180522 04:41:37<+discordbot> But it would be such a big change that I think we can't backport it to 1.14. 20180522 04:41:46<+discordbot> On top of taking quite long time to do. 20180522 04:42:38<+discordbot> (inb4 "and this wouldnt be a problem with a rolling release model because we can just push bugs and bugfixes as fast as we can without regard for packagers or translators" 20180522 04:48:55<+discordbot> Okay, I think I have the hotkeys ready 20180522 04:49:04<+discordbot> BTW, this is the commit that made SCENARIO_MUSIC and INTRO_AND_SCENARIO_MUSIC fade out the previous track: https://github.com/wesnoth/wesnoth/commit/d4e6d0bf2b9bff1c748920167eec0d0ed334364f 20180522 04:49:40<+discordbot> A UMC author can disable the fade-out by using [music] manually without specifying ms_after. 20180522 04:50:27<+discordbot> That's a bit inconvenient since it requires doing away with or overriding the stock SCENARIO_MUSIC macro. 20180522 04:50:45<+discordbot> Also, I believe that wouldn't do any changes for the [story][part] case. 20180522 04:51:49<+discordbot> Since I reckon it's the stop_music() change that's somehow involved in the transition from the main menu to the story screen. 20180522 04:52:09<+discordbot> stop_music() is called when the player turns off music in preferences. 20180522 04:53:34<+discordbot> Then why is there a fade-out before starting the story music? 20180522 04:53:57<+discordbot> prestart doesn't fire before displaying the story screen. Or does it now? 20180522 04:54:24<+discordbot> Is this comment relevant still? And if so, is it saying it’s okay to use shift, or not? -#textdomain wesnoth -# Warning: -# any hotkey that uses shift should not trigger any animation(s) in game since shift toggles accelerated mode 20180522 04:55:05<+discordbot> Holding shift still works as a temporary accel mode toggle. 20180522 04:55:13<+discordbot> Ewwww 20180522 04:55:19<+discordbot> Ew? 20180522 04:55:44<+discordbot> Well that destroys a lot of possible keystroke combinations 20180522 04:55:46<+discordbot> Isn't SCENARIO_MUSIC usually used in toplevel WML, which is run even before prestart (and indeed, before displaying the story screen)? 20180522 04:56:16<+discordbot> I mean, probably your average player wouldn’t notice a difference if holding down shift for a couple milliseconds 20180522 04:56:20<+discordbot> If you check the WML in that commit... 20180522 04:56:49<+discordbot> Also that is an incorrect assertion ancestral. 20180522 04:57:08<+discordbot> Accelerated mode can set the speed factor from anywhere between 0.25x (or 0.5x, I don't remember) to 16x. 20180522 04:57:15<+discordbot> Ah, right. The macros themselves inject prestart events. 20180522 04:57:27<+discordbot> So if the value is extreme enough, that single stroke can make a huge difference. 20180522 04:57:51<+discordbot> Can we change the acceleration speed hotkey to something else? 20180522 04:57:51<+discordbot> Whether it matters is a different story altoghet. 20180522 04:57:53<+discordbot> Caps lock? 20180522 04:58:00<+discordbot> Or just some other key 20180522 04:58:11<+discordbot> Maybe not caps lock, as you could want to type things 20180522 04:58:22<+discordbot> Ctrl? Alt? Same issue. 20180522 04:58:31<+discordbot> Really? 20180522 04:58:38<+discordbot> Ctrl and Alt? 20180522 04:58:51<+discordbot> No. 20180522 04:59:00<+discordbot> But they are used by other shortcuts. 20180522 04:59:03<+discordbot> Fading out the scenario music when the storyscreen is shown was done in a distinct commit: https://github.com/wesnoth/wesnoth/pull/719/commits/f87f09a341f967466725174f9118b3f51c6b7206 20180522 04:59:16<+discordbot> As for the obvious third alternative, Ctrl+Alt, it's a can of worms we shouldn't touch. 20180522 04:59:37<+discordbot> Several platforms like to steal events with Ctrl+Alt from applications. 20180522 04:59:51<+discordbot> Can it be a key that’s not a modifier key? 20180522 05:00:12<+discordbot> Also, on some keyboard layouts (including Finnish), right Alt actually produces Ctrl+Alt. 20180522 05:00:36<+discordbot> It's called Alt Gr on the Spanish and Latin American layouts. 20180522 05:00:43<+discordbot> Same here. 20180522 05:00:50<+discordbot> Most applications do represent it as Right Alt though. 20180522 05:01:10<+discordbot> Which is funny because it doesn't work as such in the end. 20180522 05:02:01<+discordbot> I'd suggest Scroll Lock (a.k.a. that key that I've never really understood why it even exists in the first place) but it's supposed to be a toggle. 20180522 05:02:21<+discordbot> Even though Linux or X11 (not sure which) doesn't seem to honour that and it treats it as yet another key. 20180522 05:02:27<+discordbot> Scroll Lock isn’t found on many notebook keyboards 20180522 05:02:47<+discordbot> I'm pretty sure it's usually still present but guarded behind Fn? 20180522 05:02:53<+discordbot> (Then again, maybe it’s not super important for most players to have access to?) 20180522 05:04:20<+discordbot> Scroll Lock on keyboards without a Scroll Lock key[edit] Some laptops and keyboards do not have a physical Scroll Lock key.[14] Some of these have a second function on another key that acts as if a Scroll Lock key was pressed. Some common methods are: Fn+S on certain Dell laptops.[15] Fn+C or Fn+K on certain Lenovo laptops. 20180522 05:05:05<+discordbot> I’m okay with this as long as it frees up Shift for various commands 20180522 05:10:13<+discordbot> music_thinker::process 20180522 05:10:18<+discordbot> is the relevant function 20180522 05:12:07<+discordbot> Note that fadingout_time is ms_after of the next track (defaults to zero if not specified). 20180522 05:14:26<+discordbot> i don't see why using OpenAL would be a massive refactor 20180522 05:14:31<+discordbot> the sound code is pretty contained 20180522 05:14:57<+discordbot> Not refactor. Worse, a rewrite. 20180522 05:15:21<+discordbot> And the main problem is that the risk of regressions would be very high. 20180522 05:15:39<+discordbot> When you rewrite code, there is a high risk that some corner case you don't think of changes. 20180522 05:17:16<+discordbot> I suppose 20180522 05:19:05<+discordbot> http://i.lvme.me/mil37z5.jpg 20180522 05:19:40<+discordbot> MEME 👏 REVIEW 👏 20180522 05:21:37<+discordbot> I review this meme 👌 20180522 05:23:49<+discordbot> @jyrkive Apparently the way to get through to Vultraz is by memes. 20180522 05:28:08<+discordbot> @shadowm Where is accelerated mode defined? I don’t see it in hotkeys.cfg. 20180522 05:28:28<+discordbot> it's hardcoded somewhere 20180522 05:28:37<+discordbot> Fantastic. 20180522 05:29:23<+discordbot> double display::turbo_speed() const { bool res = turbo_; if(keys_[SDLK_LSHIFT] || keys_[SDLK_RSHIFT]) { res = !res; } res |= screen_.faked(); if (res) return turbo_speed_; else return 1.0; } 20180522 05:30:02<+discordbot> I guess the question is, is it easy to move this to a definable key in hotkeys? 20180522 05:30:46<+discordbot> absolutely not 20180522 05:31:28<+discordbot> Objections if I change this key that’s hardcoded there? 20180522 05:32:02<+discordbot> (FWIW I never new this key for accelerated mode ever existed until today.) 20180522 05:32:07<+discordbot> nope 20180522 05:32:27<+discordbot> Is this documented anywhere, btw? 20180522 05:32:38<+discordbot> Like, Help or Advanced Options or something? 20180522 05:33:09<+discordbot> The latter isn't documentation. 20180522 05:33:13<+discordbot> Wait what’s this… 20180522 05:33:29<+discordbot> no idea 20180522 05:33:33<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448357731225829376/Screen_Shot_2018-05-22_at_12.33.14_AM.png 20180522 05:33:44<+discordbot> It's the permanent toggle-. 20180522 05:34:08<+discordbot> So that doesn’t mean it’s definable in hotkeys, it’s just user-definable in one of the user prefs? 20180522 05:34:31<+discordbot> I don't really see how that follows from what you posted. 20180522 05:34:35<+discordbot> it is definable is hotkeys 20180522 05:34:43<+discordbot> it's right there 20180522 05:34:44<+discordbot> line 50 20180522 05:34:59<+discordbot> We were talking about the key-hold temporary toggle, not the permanent toggle, weren't we? 20180522 05:35:06<+discordbot> Ohhhh 20180522 05:35:11<+discordbot> Perma 20180522 05:35:47<+discordbot> Permanent being, on/off switch 20180522 05:35:51<+discordbot> The latter toggles the Accelerated Mode option in Preferences, the former doesn't and only has an effect for as long as the key is held. 20180522 05:35:52<+discordbot> yes 20180522 05:36:29<+discordbot> and the permanent toggle's default is in hokeys.cfg 20180522 05:38:37<+discordbot> But you think temporary acceleration is something worth keeping? 20180522 05:38:59<+discordbot> just leave it as it is 20180522 05:39:11<+discordbot> Yes, it is worth keeping. 20180522 05:39:13<+discordbot> That means I can’t use keystrokes with Shift keys 20180522 05:39:20<+discordbot> It's extremely useful when testing animations. 20180522 05:39:21<+discordbot> uhhhh... no? 20180522 05:39:38<+discordbot> multiple keys DO use shift 20180522 05:40:01<+discordbot> As shadowm said before, if accelerated is at 16x, then… fun stuff happens? 20180522 05:40:23<+discordbot> maybe 20180522 05:40:34<+discordbot> no idea 20180522 05:40:39<+discordbot> it doesn't really matter 20180522 05:40:52<+discordbot> If you press shift it would speed things up 20180522 05:40:57<+discordbot> overthinking this, you are 20180522 05:41:21<+discordbot> shadowm just got done talking with me why it’s a bad idea, and I agree with her. 20180522 05:41:36<+discordbot> ...then the existing keys that use shift are also a bad idea 20180522 05:41:38<+discordbot> pick one 20180522 05:42:50<+discordbot> As far as I can tell, the only ones that use shift do not have any use during animated sequences. 20180522 05:43:11<+discordbot> Three of them are debug mode-only commands that can only be used while idle. 20180522 05:43:46<+discordbot> I don't know what Hold Position does but it's probably used in Planning Mode or something, also while idle. 20180522 05:44:09<+discordbot> Two of them are editor actions. 20180522 05:44:22<+discordbot> I've noticed NO weirdness in my years of playing wesnoth 20180522 05:44:25<+discordbot> so if it does happen, it's very rare 20180522 05:44:26<+discordbot> One is "Update Shroud Now" which I presume has something to do with DSU. 20180522 05:44:46<+discordbot> If it is related to DSU then it also requires the game to be idle. 20180522 05:45:39<+discordbot> None of them trigger animations or can be used during animations. 20180522 05:45:51<+discordbot> Well, except for the debug mode commands. 20180522 05:45:52<+discordbot> Okay, so should I leave it alone knowing I’m going to have a few commands with shift? 20180522 05:45:57<+discordbot> But they are debug mode commands. 20180522 05:46:07<+discordbot> good god don't buy fruity cheerios 20180522 05:46:09<+discordbot> they taste foul 20180522 05:46:25<+discordbot> Now if you decided that Recruit should be Shift+R (dumb example but it works), that'd be a problem. 20180522 05:46:35<+discordbot> Or especially Repeat Recruit. 20180522 05:46:53<+discordbot> Both play an animation (the unit's recruit animation). 20180522 05:47:05<+discordbot> http://mproud.com/wesnoth/shortcuts/shortcuts.png 20180522 05:47:21<+discordbot> I haven’t touched that in 3 years 20180522 05:47:24<+discordbot> Speak to Ally can happen during animations. 20180522 05:47:47<+discordbot> if these are the changes you'll be making 20180522 05:47:57<+discordbot> i shall include it in the announcement 20180522 05:48:00<+discordbot> And so can Planning Mode or Zoom In/Out. 20180522 05:48:01<+discordbot> The original key will still work except were * are 20180522 05:48:23<+discordbot> Well 20180522 05:48:25<+discordbot> Currently 20180522 05:48:30<+discordbot> Zoom in is +, right? 20180522 05:48:37<+discordbot> yes 20180522 05:48:56<+discordbot> On U.S. keyboards without using the keypad, that requires the use of the shift key 20180522 05:49:04<+discordbot> shrugs 20180522 05:49:14<+discordbot> Anyway I did say "whether it matters is a different story altogether". 20180522 05:49:15<+discordbot> I've had no prblems 20180522 05:49:34<+discordbot> Although you should probably keep in mind that not all people will take the same amount of time to press Shift+M to speak. 20180522 05:49:48<+discordbot> I have no idea if shift + = as "+" means something different than the combination of shift and = 20180522 05:49:54<+discordbot> Some people may hold Shift for longer while they coordinate movement to press the modified key. 20180522 05:50:16<+discordbot> I can change Shift M and shift P to Alt or something else 20180522 05:50:18<+discordbot> Not all of us have the same typing skill or motor dexterity. 20180522 05:50:20<+discordbot> Or not change it 20180522 05:50:26<+discordbot> plain M should be message 20180522 05:50:38<+discordbot> Plain M still is 20180522 05:51:10<+discordbot> dota uses Enter and Shift + Enter for chat and ally chat, respectively 20180522 05:51:34<+discordbot> In my head, I assumed the natural progression of familiarity with modifier keys are shift, then ctrl, then alt 20180522 05:51:54<+discordbot> But I don’t think it matters much 20180522 05:52:43<+discordbot> In Civ, when you press Return, it cycles to the next unit, doesn’t it? 20180522 05:53:15<+discordbot> (until there’s nothing else to do and then it ends the turn? I may be misremembering.) 20180522 05:53:42<+discordbot> i don't play civ 20180522 05:55:04<+discordbot> Alright, so maybe I’ll change Speak to Ally and Planning Mode to Alt 20180522 05:55:13<+discordbot> And leave everything else 20180522 05:55:19<+discordbot> And Planning Mode is no longer P 20180522 05:55:41<+discordbot> (And me from 5 years ago rejoices!) 20180522 05:55:58<+discordbot> (I was so worked up and bothered by that.) 20180522 05:57:34<+discordbot> Hmm, Statistics/Status Table, S and T… 20180522 05:58:02<+discordbot> Maybe I should flip that 20180522 07:14:17< Rhonda> Hmm, guess I have to special-case something for mips/mipsel because it runs into memory exhausted. :) 20180522 07:15:31-!- irker078 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout] 20180522 07:16:12<+discordbot> You guys/girls should already have a way to deal with exceptionally large programs there, right? 20180522 07:16:40<+discordbot> I don't think Wesnoth is the single Debian package that requires the highest amount of memory to compile. 😉 20180522 07:31:14-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20180522 07:36:39<+discordbot> idly wonders for the hundredth time why the standard library doesn't have its own version of boost::adaptors::reverse 20180522 07:38:50<+discordbot> It's a part of the range proposal. 20180522 07:38:52<+discordbot> http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2014/n4128.html 20180522 07:39:56<+discordbot> In Boost, boost::adaptors::reverse is part of Boost.Range. It depends on the rest of the range infrastructure. 20180522 07:43:50<+discordbot> oh. 20180522 07:43:50<+discordbot> huh 20180522 07:46:19<+discordbot> and they haven't decided what to do with that? 20180522 07:47:13<+discordbot> It's still not in the standards track. 20180522 07:47:24<+discordbot> At best it might arrive in early 2020s. 20180522 07:47:30<+discordbot> long time, these people take 20180522 07:48:19<+discordbot> anyway 20180522 07:48:31<+discordbot> I'm making my first bit of progress on a_r in awhile 20180522 07:48:56<+discordbot> I've gotten rid of the individual canvas textures like you asked 20180522 07:49:17<+discordbot> Nice. 😃 20180522 07:55:24< Rhonda> jyrkive: (btw., using "folks" instead of "guys/girls" is more approachable and less bulky :) 20180522 07:55:49<+discordbot> Thanks, noted. (I'm not a native English speaker.) 20180522 07:57:01< Rhonda> wesnoth is definitely at the top of ressources needed, not sure how libreoffice or iceweasel/chromium deal with it, they potentially also should run in such issues. 20180522 07:57:24< Rhonda> It was suggested to me to reduce the debug symbols built on those platforms though. 20180522 07:58:00<+discordbot> Just "you" is also acceptable, as it's both singular and plural. 20180522 07:58:22<+discordbot> I know, but I wanted to highlight that I meant the plural form. 20180522 08:01:43<+discordbot> I usually use "you guys" but I suppose someone would say that's overly gendered language that one should avoid 20180522 08:02:19<+discordbot> I considered it and decided not to use it. 20180522 08:03:48<+discordbot> (I don't particularly understand the idea that gender should be stripped from colloquial language, but... well, I'm not someone whos affected by it) 20180522 08:04:05<+discordbot> You can always go full American and say "Ya'll" 😛 20180522 08:08:05< Rhonda> That's why I pointed at "folks" instead. :) 20180522 08:08:53< zookeeper> you didn't need to port my commits, i didn't do it simply because it was obnoxiously late and i had to get sleep ASAP, and i would have done it myself... well, by now probably. 20180522 08:09:57<+discordbot> wait 20180522 08:09:59<+discordbot> .... 20180522 08:10:03<+discordbot> what the hell 20180522 08:10:06<+discordbot> all of the text problems 20180522 08:10:08<+discordbot> are magically fixed 20180522 08:10:12<+discordbot> O_O 20180522 08:10:18<+discordbot> In master? 20180522 08:10:31<+discordbot> by changing to direct-to-screen rendering 20180522 08:12:37<+discordbot> I'm not surprised in the least. 20180522 08:12:50< zookeeper> @Vultraz, no, the Ds is some garbage 20180522 08:13:15<+discordbot> The whole issue is that some information was lost in the process of converting Pango renders to SDL surfaces. 20180522 08:13:40<+discordbot> Particularly because of the premultiplied alpha -> discrete alpha conversion. 20180522 08:13:45< zookeeper> some of the terrain codes were just... absolutely nonsensical. sunken castle, a non-existant code like Uui, etc. 20180522 08:14:13<+discordbot> Could you remove it, then 20180522 08:16:50< zookeeper> P.S. if someone wants to playtest that scenario (SoF finale), then... uh, wouldn't hurt. i'm pretty sure i tried all routes of finishing the scenario, but still. 20180522 08:17:32< zookeeper> sure, i'll remove it. first i'll check why there was a conflict in the first place... 20180522 08:18:07<+discordbot> please do 20180522 08:18:23<+discordbot> @jyrkive I'm not sure if I still need the shape cache... I think not? 20180522 08:18:55< zookeeper> oh, right, celmin ran 05d6c842 (schema validator fixes) on 1.14 only. 20180522 08:19:41< zookeeper> or rather, hasn't forward-ported it yet, over a month later. maybe you shouldn't complain to _me_ about slowness... :p 20180522 08:20:02<+discordbot> but...I thought he already... 20180522 08:20:04<+discordbot> BLAH 20180522 08:20:21<+discordbot> Indeed, the shape cache isn't needed either. 20180522 08:24:11-!- irker017 [~irker@uruz.ai0867.net] has joined #wesnoth-dev 20180522 08:24:11< irker017> wesnoth: ln-zookeeper wesnoth:master cdea2e70cfa4 / data/campaigns/Sceptre_of_Fire/scenarios/9_Caverns_of_Flame.cfg: SoF S9: Removed garbage terrain code https://github.com/wesnoth/wesnoth/commit/cdea2e70cfa446322e59174e28e1ca74183c5f4b 20180522 08:26:20<+discordbot> ty 20180522 08:38:24< irker017> wesnoth: Charles Dang wesnoth:master 02858f225933 / src/ (7 files in 4 dirs): GUI2/Canvas: discard individual canvas texture in favor of direct-to-screen rend https://github.com/wesnoth/wesnoth/commit/02858f225933ba56c088ab39fd3231424803b5ba 20180522 08:47:11<+discordbot> @jyrkive so, the textures are still fragmented in the cache (both image and text). But we can address that later 20180522 08:47:27<+discordbot> that's basically the last step 20180522 08:48:23<+discordbot> I really need to tackle that event problem... 20180522 08:50:11<+discordbot> What do you mean with "fragmented"? 20180522 08:50:19<+discordbot> 1 texture per image 20180522 08:50:34<+discordbot> it's the same mechanism as the surface cache 20180522 08:50:36<+discordbot> just with textures 20180522 08:50:42<+discordbot> Addressing that should be the first step, not last. 20180522 08:50:53<+discordbot> It requires massive refactoring of the rendering pipeline. 20180522 08:51:37<+discordbot> well, then I need to get my spritesheet generator fully working so we can start having the game load the pre-rendered sheets. 20180522 08:52:38<+discordbot> I had been experimenting with data output in the generator, so we have files that look like this: [image] path="terrain/cave/earthy-floor-s-sw.png" rect="685,0,63,47" [/image] [image] path="terrain/cave/earthy-floor-n.png" rect="748,0,43,18" [/image] [image] path="terrain/cave/earthy-floor-ne-se.png" rect="791,0,38,72" [/image] 20180522 08:52:41<+discordbot> I'd need to finish the low-level classes, as well as improve the spritesheet generator. 20180522 08:52:56<+discordbot> But as long as we're making 1.14.x releases, they take priority. 20180522 08:53:04<+discordbot> the idea being that the individual paths would map to rects in the sheet 20180522 08:53:50<+discordbot> You also need to store how much transparent area there was and how the remaining part is anchored relative to that. 20180522 08:54:04<+discordbot> Remember that your spritesheet generator is removing transparent areas. 20180522 08:54:29<+discordbot> hmmm 20180522 08:54:33<+discordbot> do we really? 20180522 08:54:44<+discordbot> Yes. Absolutely. It's not optional. 20180522 08:54:50<+discordbot> why 20180522 08:55:30<+discordbot> we could easily make it so image render in the right locations without excess alpha. 20180522 08:55:32<+discordbot> Suppose, for example, that an image is supposed to be anchored to center (what my sprite class is assuming, but probably not true for all images). 20180522 08:56:19<+discordbot> If a sprite has, say, bottom half fully transparent and your spritesheet generator removes that, the remaining part should now be anchored to bottom. 20180522 08:57:25<+discordbot> And that's an exceptionally simple case. It gets ten times more complicated when sprites have varying amounts of transparency in all edges, and the reference anchor varies (isn't guaranteed to be the center). 20180522 08:58:06<+discordbot> hmm... 20180522 08:59:13<+discordbot> well, next thing i should do is wire the sheet generator in as a command line option 20180522 08:59:26<+discordbot> so it's easier to run and test 20180522 09:00:02<+discordbot> plus, that would be the preferred method by which players will generate their own sheets 20180522 09:03:54-!- vladimirslavik [vslavik@nat/redhat/x-fssobytltlplfbiy] has quit [Quit: Leaving] 20180522 09:25:28-!- gfgtdf [~gfgtdf@134.76.63.8] has joined #wesnoth-dev 20180522 09:26:33< irker017> wesnoth/wesnoth:1.14 Iris Morelle d2e94b320f Do not allow [message] to scroll to unit AppVeyor: All builds passed 20180522 09:41:04<+discordbot> HM 20180522 09:41:08<+discordbot> oops 20180522 09:41:22<+discordbot> I discarded something Iwasn't supposed to discard 20180522 09:41:25<+discordbot> 😬 20180522 09:43:27<+discordbot> or.. 20180522 09:43:27<+discordbot> hm.. 20180522 09:45:43<+discordbot> ah 20180522 09:47:24<+discordbot> So RIP work? 😄 20180522 09:47:55<+discordbot> nah 20180522 09:48:12<+discordbot> I just removed the "don't remove immutable shapes" thing from clear_shapes 20180522 09:48:14<+discordbot> simple re-add 20180522 09:48:27<+discordbot> 😄 20180522 09:50:23<+discordbot> Shouldn't be new version tagged already? 20180522 09:50:23<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448422362279444481/unknown.png 20180522 09:50:41<+discordbot> I delayed it since I want @ancestral 's hotkey PR 20180522 09:50:54<+discordbot> oh, ok. np 20180522 09:51:23<+discordbot> soon as it's in I'll tag 20180522 10:04:40< irker017> wesnoth: Charles Dang wesnoth:master eecc46f02f9b / src/gui/core/ (canvas.cpp canvas.hpp): GUI2/Canvas: restored handling of immutable shapes (fixup 02858f2) https://github.com/wesnoth/wesnoth/commit/eecc46f02f9b65236674d27c5786b12cb729c7ea 20180522 10:29:45<+discordbot> Sorry about those indents, @Vultraz. Pretty sure I've caught and fixed them all now. 20180522 10:30:03<+discordbot> np 20180522 10:35:44< irker017> wesnoth: Charles Dang wesnoth:master b2fc0f25e0e8 / src/gui/widgets/ (29 files): GUI2: removed unused offset parameters from drawing functions https://github.com/wesnoth/wesnoth/commit/b2fc0f25e0e8e3718612d596023f87058eb8a8b0 20180522 10:55:06< Ravana_> that hotkey image says ctrl+m and shift+m are speak to ally. m and tab is speak. but what is speak to all? 20180522 11:15:32< irker017> wesnoth: Charles Dang wesnoth:master 2819f549a1a6 / src/gui/ (core/canvas.cpp core/canvas.hpp widgets/minimap.cpp widgets/minimap.hpp): GUI2:Canvas: remove custom drawing function interface https://github.com/wesnoth/wesnoth/commit/2819f549a1a6a02f629a557ba84f73d213f2e2a8 20180522 11:15:35< irker017> wesnoth: Charles Dang wesnoth:master 1a9c6a7eca4c / src/gui/widgets/widget.cpp: Removed an outdated comment https://github.com/wesnoth/wesnoth/commit/1a9c6a7eca4c98432dcf3e8f0d152ca4cd429d25 20180522 11:16:08<+discordbot> It's quite interesting how much overhead was necessary to maintain the one-texture-per-canvas drawing method 20180522 11:18:06<+discordbot> canvas::draw is now 16 lines long 20180522 11:21:32< irker017> wesnoth: Charles Dang wesnoth:master 6e4327f313b2 / src/gui/core/canvas.cpp: GUI2/Canvas: reset size_changed_ flag after setting formula variables (fixup 028 https://github.com/wesnoth/wesnoth/commit/6e4327f313b295e54dec566ea9e3fa65ae34f71e 20180522 11:21:41<+discordbot> 18* 😛 20180522 11:30:16<+discordbot> @jyrkive out of curiosity, how would the blur shader be applied to individual UI elements? 20180522 11:30:58<+discordbot> Each draw_op would contain a reference to the correct shader. 20180522 11:31:33<+discordbot> The rendering pipeline would sort draw operations by shader, then activate the right shader before sending corresponding draw operations to the GPU. 20180522 11:33:50<+discordbot> hm, I see. 20180522 11:33:58<+discordbot> this is indeed actually low-priority 20180522 11:34:01<+discordbot> (UI blurring) 20180522 11:34:38<+discordbot> It's important to get the rendering to actually work. Shader support can be added much later. 20180522 11:34:59-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009] 20180522 11:35:04<+discordbot> UI rendering works completely 20180522 11:35:15<+discordbot> (except for the credits) 20180522 11:35:37<+discordbot> ...with the horrible, horrible "one texture per image" implementation. 20180522 11:36:01<+discordbot> Yes, but that's a backend problem. 20180522 11:36:29<+discordbot> it can be solved. 20180522 11:36:56<+discordbot> need to start somewhere 20180522 11:37:34<+discordbot> the texture cache I added was only intended to be a stopgap fix 20180522 11:37:56<+discordbot> to the extent that it would at least prevent constant disk polling 20180522 11:38:20<+discordbot> also I had no idea how to write anything better 😛 20180522 11:39:43<+discordbot> I'll leave that area to you. 20180522 11:40:14<+discordbot> Sure. But like I said, I'll focus on 1.14.x releases as long as they are being made. 20180522 11:40:52<+discordbot> They'll be made as long as there's work being done 20180522 11:41:01<+discordbot> Bit circular 😬 20180522 11:41:43< irker017> wesnoth: Charles Dang wesnoth:master b011faa24481 / src/gui/core/ (canvas.cpp canvas.hpp): GUI2/Canvas: merge draw() and render() and remove commented-out blur code https://github.com/wesnoth/wesnoth/commit/b011faa24481234f561ab5fa51018ee5e9359f6e 20180522 11:42:09<+discordbot> Solving that is your job as the project manager. 20180522 11:42:44<+discordbot> You have set up a situation where no one really wants to work on master. Most people can't, and even those who can, can achieve much faster results in 1.14.x. 20180522 11:43:58<+discordbot> To be fair, you encouraged me to merge a_r quickly. 20180522 11:44:08<+discordbot> But that's beside the point. 20180522 11:44:19<+discordbot> It's really only you and me doing deep engine work. 20180522 11:44:19<+discordbot> Sorry. I didn't know it was in an unplayable state. 😦 20180522 11:45:00<+discordbot> My logic was that merging it would get people working on it, but I neglected the fact that... we don't really have anyone else to work on it 20180522 11:46:04<+discordbot> In addition, no one wants to keep fixing stuff someone else has broken (or worse, still keeps breaking). 20180522 11:46:39<+discordbot> There are like, 5 of us. gfgtdf is fixing the occasional gamestate bug, which can be don on 1.14. celmin is working on the WML engine, also can be done on 1.14. shadowm is doing minor fixes which can be done on 1.14. So it's only you and me in any position to do anything with master and a_r 20180522 11:46:53<+discordbot> "I did the easy part: now someone else do the hard part!" would be an outright arrogant attitude (not that you were actually doing that). 20180522 11:50:28-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev 20180522 11:55:04<+discordbot> I didn't want someone else to do the work, I wanted other people to be able to help with the work 20180522 11:55:38<+discordbot> Well, the hardest part of refactoring is fixing the resulting bugs. 20180522 11:55:51<+discordbot> Bug fixing is always the most difficult part of programming. 20180522 12:15:40< irker017> wesnoth/wesnoth:master ln-zookeeper 59b69288ba Updated changelogs AppVeyor: All builds passed 20180522 12:39:58< zookeeper> nothing wrong with people primarily working on 1.14 for now. it's new, will be the current stable for a long time, has bugs, needs work, etc. 20180522 13:02:59<+discordbot> Yes 20180522 13:03:08<+discordbot> perhaps we can do campaign work on 1.14 and engine work on master 20180522 13:29:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180522 13:43:39-!- travis-ci [~travis-ci@ec2-54-166-169-32.compute-1.amazonaws.com] has joined #wesnoth-dev 20180522 13:43:40< travis-ci> wesnoth/wesnoth#18317 (master - b011faa : Charles Dang): The build passed. 20180522 13:43:40< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/382116227 20180522 13:43:40-!- travis-ci [~travis-ci@ec2-54-166-169-32.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180522 14:30:10-!- gfgtdf [~gfgtdf@134.76.63.8] has quit [Ping timeout: 260 seconds] 20180522 14:34:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180522 14:36:19-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180522 14:38:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180522 14:42:55< irker017> wesnoth/wesnoth:master Charles Dang 02858f2259 GUI2/Canvas: discard individual canvas t AppVeyor: All builds passed 20180522 15:14:23-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180522 15:14:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180522 16:16:41<+discordbot> I can’t figure out why alt+return doesn’t work 20180522 16:31:35< Ravana_> with 1.14 ctrl+return no longer behaves as ctrl+j, and not possible to assign to anything either 20180522 16:37:40< irker017> wesnoth/wesnoth:master UnwiseOwl 671ac7e4dc fixed more indents AppVeyor: All builds passed 20180522 16:40:32<+discordbot> Okay figured it out. Depending where the commands are defined in the file is important. 20180522 16:40:50<+discordbot> If hotkey return is defined before alt+return, then the second will not get looked at 20180522 16:41:02<+discordbot> (unless maybe it has alt=no) 20180522 16:41:40<+discordbot> Ravana_: Not sure what you mean 20180522 16:41:56<+discordbot> Do you want it to work like ctrl+j? 20180522 16:42:19< Ravana_> yes 20180522 16:47:12<+discordbot> I mean, I can do that 20180522 16:49:08<+discordbot> Hey, how do I bring up the lua console? 20180522 16:49:23<+discordbot> I press ` and nothing happens 20180522 16:50:29<+discordbot> ` + space, or `` + space 20180522 16:50:53<+discordbot> However, that's because it is a dead key 20180522 16:50:59<+discordbot> Should just be a tick 20180522 16:51:02<+discordbot> What I mean in 20180522 16:51:15<+discordbot> What should happen? 20180522 16:51:32<+discordbot> Because I feel like nothing is happening 20180522 16:55:18<+discordbot> If I want to type è I type effectifely (shift + [or shift + ´ to be precise], - nothing displays yet because it is a dead key - and when typing the next letter it is used, usually to bring up an modified letter. When the next letter is the same one, or space, then prints for real - and this is the shortcut. 20180522 16:55:28<+discordbot> crap, discord makrup ... 20180522 16:56:13<+discordbot> You can write ``` to prevent it from being detected as markup. 20180522 16:56:52<+discordbot> Thanks 20180522 17:00:59<+discordbot> Ignore the part with the space - that works on a text field such as here, but to open the console I have to type twice `` [or shift + ´ + shift + ´ to be precise] 20180522 17:11:53<+discordbot> How do I bring up the lua console? 20180522 17:21:14-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20180522 17:21:24< Ravana_> I assigned lua console to ' 20180522 17:21:49<+discordbot> You need to be in debug mode. 20180522 17:25:36<+discordbot> Regarding the memory leak in that large map: I can't reproduce any leak when just scrolling around. Memory usage settles to 1113 MB and stops increasing. 20180522 17:26:06<+discordbot> However. It appears that we leak three megabytes every time the player views Unit Description of any unit and closes it. 20180522 17:27:14<+discordbot> Okay, thank you 20180522 17:27:36-!- gfgtdf [~gfgtdf@134.76.63.8] has joined #wesnoth-dev 20180522 17:49:39< irker017> wesnoth/wesnoth:master UnwiseOwl cefc0801f6 fixed more indents AppVeyor: All builds passed 20180522 18:01:43<+discordbot> Hmm. Apparently standardize_surface_format() is too aggressive and ends up standardizing the image format of every image we load. Making it less aggressive may well shorten loading times notably... 20180522 18:02:29<+discordbot> Have you considered excluding JPEG images from being cached btw? 20180522 18:03:15<+discordbot> No. I think it doesn't matter much whether we cache them or not. 20180522 18:03:42<+discordbot> They tend to be invariably large though. 20180522 18:04:24<+discordbot> And (hopefully, too lazy to check the add-ons server myself) rarely used with image path functions. 20180522 18:09:12<+discordbot> What’s the difference between speak and speaktoall? 20180522 18:09:31< Ravana_> speak preserves last setting 20180522 18:09:43< Ravana_> speaktoall and speaktoally set setting 20180522 18:09:57<+discordbot> “Last setting?” 20180522 18:10:05< Ravana_> team chat 20180522 18:10:06<+discordbot> Oh 20180522 18:10:10<+discordbot> I see 20180522 18:10:23<+discordbot> So m should probably be speak 20180522 18:14:21<+discordbot> Found a potential reason why I can't reproduce the memory leak. 20180522 18:14:59<+discordbot> I'm using SDL 2.0.4. With such an old version, surface::free_surface() calls SDL_FreeSurface() unconditionally. 20180522 18:15:27<+discordbot> Starting from SDL 2.0.6, it only decrements the reference count if the refcount is more than one. 20180522 18:33:25< irker017> wesnoth: Severin Glöckner wesnoth:1.14 2e04dde374c8 / po/ (5 files in 5 dirs): updated German translation https://github.com/wesnoth/wesnoth/commit/2e04dde374c8fd4ec797a2bcdaf60ad4c48aaafd 20180522 18:37:29<+discordbot> Ivanovic: Can you remind me what the frequency of pot updates on master before the first development release is supposed to be, as well as what happens to new/modified strings from a stable branch po merged into an outdated dev branch pot? 20180522 18:38:01<+discordbot> (IOW what happens if the po contains msgids that can't be found in the pot that's being merged into it.) 20180522 18:38:19< Ivanovic> 1) pot-update for dev should happen "shortly before" the first release, especially when translators are supposed to still focus on stable 20180522 18:38:26< Ivanovic> 2) I just apply the language files against both pot files 20180522 18:38:48< Ivanovic> this way the "new" strings are already in the file in the commented out section of dev 20180522 18:39:00< Ivanovic> yes, this marks more strings than necessary as fuzzy, but who cares? 20180522 18:39:01<+discordbot> Ahhhhhhh, okay, so they're not lost. 20180522 18:39:25< Ivanovic> if you want to you can also move over the po files from stable to dev before running the very first pot-update 20180522 18:39:48< Ivanovic> only the changed strings will be marked as fuzzy and stay fuzzy 20180522 18:52:28-!- Holius [5a5b9859@gateway/web/freenode/ip.90.91.152.89] has joined #wesnoth-dev 20180522 19:04:50< irker017> wesnoth/wesnoth:master UnwiseOwl 7dc5254470 removed empty line AppVeyor: All builds passed 20180522 19:06:29<+discordbot> Hmm, looks like GUI1 text cache (50 entries) is too small, or some kind of bug blocks it from working correctly. 20180522 19:07:10<+discordbot> When I view the Spearman unit description, none of the strings in it are in the cache, even right after closing the same unit description page. 20180522 19:15:36<+discordbot> Okay, the problem is that it's too small. Most text in it is stuff from the help sidebar. One string per opened topic, which easily becomes dozens of strings when viewing unit descriptions. 20180522 19:24:12<+discordbot> https://github.com/SDL-mirror/SDL/commit/0506d49a06e3feb23aadb41d6e4a2d0f43eedda6 20180522 19:25:05<+discordbot> Turns out that the SDL bug that required the "don't call SDL_FreeSurface() for surfaces which have other references" workaround was fixed in SDL 2.0.7. 20180522 19:25:49<+discordbot> Assuming that the workaround is what causes the memory leak, I could simply disable it with all SDL versions except 2.0.6. 20180522 19:26:28<+discordbot> SDL 2.0.6 users would still be affected by the memory leak, but at least most users would get rid of it. 20180522 19:55:38<+discordbot> what version is used by the official builds? 20180522 19:57:16<+discordbot> I think it's 2.0.8. 20180522 20:02:50<+discordbot> Good evening, I'm the guy who started working on description for the Quenoth Elves. As you can see, I haven't been active since a long time. I've had... other things to worry about, but I haven't forgotten and I will finish it when I get a bit more time and the will to work properly 20180522 20:11:14< zookeeper> @MarkoLeSerbe, no worries. 20180522 20:11:59<+discordbot> OK, turns out that I can reproduce the memory leak on GNU/Linux (with SDL 2.0.8). 20180522 20:12:06< zookeeper> uh, so... https://forums.wesnoth.org/viewtopic.php?f=39&t=8715&p=628621#p628619 <- what's up with that thing about getting taken to the credits? has _that_ been fixed? 20180522 20:12:36<+discordbot> Unfortunately I can still reproduce it even if I remove the "avoid calling SDL_FreeSurface() workaround. 20180522 20:13:06<+discordbot> The memory leak must be caused by something else. I'll investigate later. 20180522 20:38:47-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180522 20:39:00-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180522 20:49:22-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 245 seconds] 20180522 21:27:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180522 21:27:29-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180522 21:59:47-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180522 22:00:02-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180522 22:01:06-!- Holius [5a5b9859@gateway/web/freenode/ip.90.91.152.89] has quit [Quit: Page closed] 20180522 22:04:30< irker017> wesnoth/wesnoth:master UnwiseOwl a3af3e3e92 correct indent AppVeyor: All builds passed 20180522 22:11:52-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180522 22:12:04-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180522 22:37:09<+discordbot> @ancestral confuses me, your PR does 20180522 22:37:33<+discordbot> Okay, let me do something different 20180522 22:37:41<+discordbot> Give me a moment 20180522 22:37:50<+discordbot> Stuff like shift = 20180522 22:37:55<+discordbot> that's... not a thing 20180522 22:37:59<+discordbot> you just need + 20180522 22:38:05<+discordbot> and why is = not zoomdefault 20180522 22:38:52<+discordbot> Look at your keyboard 20180522 22:39:01<+discordbot> Where’s your plus? 20180522 22:39:15<+discordbot> You don't need to assign a hotkey to shift = to make it work with + 20180522 22:39:18-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer] 20180522 22:39:25<+discordbot> Apparently neither works 20180522 22:39:32<+discordbot> Even though the combination is indeed "shift =" 20180522 22:39:34<+discordbot> Which confuses me 20180522 22:40:17<+discordbot> wait 20180522 22:40:18<+discordbot> .... 20180522 22:40:28<+discordbot> [hotkey] command=zoomin key="=" [/hotkey] 20180522 22:40:29<+discordbot> uhhh 20180522 22:40:32<+discordbot> (that's from 1.14) 20180522 22:40:36<+discordbot> ummm.... 20180522 22:40:49<+discordbot> whaa? 20180522 22:41:00<+discordbot> The point here is zooming shouldn’t need a shift key 20180522 22:41:15<+discordbot> ok, I guess = can be zoomin... 20180522 22:41:38<+discordbot> I’m going to push a new PR in a short bit 20180522 22:41:43<+discordbot> alright, why don't you just leave it as it is now. 0, - and = for default, zoom out, and zoom in 20180522 22:42:01<+discordbot> I can get rid of the shift variations 20180522 22:42:07<+discordbot> yeah 20180522 22:42:08<+discordbot> They don’t seem to work for me anyway 20180522 22:42:13<+discordbot> Not sure why 20180522 22:42:15<+discordbot> I was mistaken the key was + 20180522 22:54:45-!- travis-ci [~travis-ci@ec2-54-162-4-12.compute-1.amazonaws.com] has joined #wesnoth-dev 20180522 22:54:46< travis-ci> gfgtdf/wesnoth#1129 (1.14 - 89498eb : gfgtdf): The build was broken. 20180522 22:54:47< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth/builds/382401915 20180522 22:54:47-!- travis-ci [~travis-ci@ec2-54-162-4-12.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180522 23:13:50< irker017> wesnoth: ancestral wesnoth:cleanup_keys ccc223b30f7b / data/core/hotkeys.cfg: Reorganized hotkeys file so modifiers appear before non-modifiers. This prevents https://github.com/wesnoth/wesnoth/commit/ccc223b30f7bc53e90e12884627fafe29c47ed9a 20180522 23:17:37<+discordbot> @Vultraz Any idea what could possibly be going on here? https://forums.wesnoth.org/viewtopic.php?p=628633#p628633 20180522 23:18:02<+discordbot> Notice what his add-ons are displayed like. 20180522 23:18:33<+discordbot> I do not 20180522 23:18:41<+discordbot> it makes no sense 20180522 23:18:55<+discordbot> Whereas here things work fine (ignore the fact that the icon frame is empty, that's because there's an image being BLITted that no longer exists at the expected path): 20180522 23:18:56<+discordbot> https://cdn.discordapp.com/attachments/259976436490829825/448625838628405249/unknown.png 20180522 23:19:29<+discordbot> I wish you could think of something because I don't have anything else to suggest in that thread right now. 20180522 23:23:46<+discordbot> Well. 20180522 23:25:10<+discordbot> I hate Windows with a passion but I just got the RS4 update on my laptop and it's a depressing reminder of how much UI stuff we can't do in Wesnoth yet while everyone else has been able to do it for a decade. 20180522 23:28:00<+discordbot> such as? 20180522 23:28:14< gfgtdf> can someone with a debugger help me a bit currently ? (generate a stacktrace) 20180522 23:28:24<+discordbot> you mean the really cool list mouseover effect? 20180522 23:28:31<+discordbot> Yeah. 20180522 23:33:30<+discordbot> I honestly have no idea how one would even conceptually design that 20180522 23:34:33<+discordbot> Uh, the concept is simple. 20180522 23:34:44<+discordbot> Things get highlighted according to the mouse position. 20180522 23:35:00<+discordbot> It's the implementation that's decidedly non-trivial. 20180522 23:36:03<+discordbot> But you could still trivially implement it on a widget-by-widget basis. 20180522 23:37:04<+discordbot> @ancestral The reason why the zooming hotkeys don't work for you is probably https://github.com/wesnoth/wesnoth/issues/1984 20180522 23:38:21-!- travis-ci [~travis-ci@ec2-54-162-4-12.compute-1.amazonaws.com] has joined #wesnoth-dev 20180522 23:38:22< travis-ci> wesnoth/wesnoth#18322 (cleanup_keys - ccc223b : ancestral): The build passed. 20180522 23:38:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/382418233 20180522 23:38:22-!- travis-ci [~travis-ci@ec2-54-162-4-12.compute-1.amazonaws.com] has left #wesnoth-dev [] 20180522 23:40:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180522 23:42:09<+discordbot> Hm, yes, you’re right. 20180522 23:51:23<+discordbot> @sevu blah... 20180522 23:52:19-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20180522 23:52:19<+discordbot> I can't even follow hotkey issues anymore 20180522 23:52:38< celticminstrel> ??? 20180522 23:52:59<+discordbot> the system makes no sense, I can't understand the concepts, and half of what we're doing is unconventional 20180522 23:53:30<+discordbot> guess ancstral will just have to use keys that don't require special combinations 20180522 23:53:52< celticminstrel> ... 20180522 23:54:00< celticminstrel> Shift+= is totally a normal way to get + though. 20180522 23:54:09<+discordbot> yes 20180522 23:54:10< celticminstrel> How else do you expect to show it? 20180522 23:54:20<+discordbot> + 20180522 23:54:26< celticminstrel> That's ambiguous. 20180522 23:54:27<+discordbot> but it seems we're not using + 20180522 23:54:28<+discordbot> we're using = 20180522 23:54:49<+discordbot> see, that's the thing 20180522 23:54:57<+discordbot> is the hotkey the combination of shift and = 20180522 23:55:11<+discordbot> or the + "key", however that's produced 20180522 23:55:28< celticminstrel> There is a + key on the numeric keypad. 20180522 23:55:42<+discordbot> that's not my question 20180522 23:56:02< celticminstrel> Writing shift+= makes it clear which one you mean. 20180522 23:56:13< celticminstrel> Writing + does not. 20180522 23:56:40<+discordbot> are hotkey supposed to be a combination, regardless of what it produces on any keyboard, or are they supposed to be specific keys, like "+", regardless of the combinations that produce them on any keyboard 20180522 23:56:56< celticminstrel> Hotkeys are typically chords. 20180522 23:57:11< celticminstrel> That is, several keys pressed simultaneously. 20180522 23:57:40<+discordbot> right 20180522 23:57:45<+discordbot> so it seems we want + 20180522 23:58:02< celticminstrel> No? Like I said, that's ambiguous. 20180522 23:58:21<+discordbot> why does it matter if it's ambiguous 20180522 23:58:32<+discordbot> you should be able to press any + key to zoom 20180522 23:58:37< celticminstrel> No. 20180522 23:58:38<+discordbot> (in our theoretical scenario) 20180522 23:58:48< celticminstrel> There's no reason for both + keys to have the same effect. 20180522 23:59:03<+discordbot> and I bet this is why we use text input, isn't it... 20180522 23:59:05<+discordbot> 😐 --- Log closed Wed May 23 00:00:33 2018