00:00 < bridge> Up and down worked 00:00 < bridge> i tried a few months ago and the hats werent working 00:00 < bridge> It was broken in a different way in a few previous versions 00:00 < bridge> and the same joystick shenangans 00:00 < bridge> ^ whichi cant figure out why is happening 00:00 < bridge> my partner wants to get into ddnet but they prefer controller 00:00 < bridge> lmao 00:01 < bridge> I'd even consider mobile easier than controller 😄 00:02 < bridge> being able to use hats for wasd and relative movement on the right joystick 00:02 < bridge> it feels the same to me as mouse and kb 00:02 < bridge> it would take me a while to get used to it 00:02 < bridge> I'd definitely use absolute aiming though 00:04 < bridge> i use really low mouse max distance 00:04 < bridge> so to aim i move my mouse in the direction i want to aim 00:38 < bridge> render and ui related code are brrr.. 00:38 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345540858262454322/20250225_053516.jpg?ex=67c4ec13&is=67c39a93&hm=59419e6e45976c5d90efc9cdfba0e961d1cce9f8d071541c8f5b4d8c34edc765& 00:39 < bridge> Spaghetti everywhere 00:41 < bridge> I think it's just very redundant, maplayers especially. The engine part just follows usual vulkan backend api with shaders and rasterization and stuff. So at least the render related code is not _that_ aweful 00:42 < bridge> Well I mean more specific code like how you actually render the tees 00:44 < bridge> Me trying to get alpha thing work in demo 00:46 < bridge> I am wondering, are there any rules against functions as parameters? If not I'll follow up with a refactoring ... 00:46 < bridge> I love recalls 00:47 < bridge> I love callbacks 01:02 < bridge> @everyone https://discord.gg/9N9N9abqVm 02:14 < bridge> ok tune lock tile is ready ill just wait until #9788 is merged 😄 02:14 < bridge> https://github.com/ddnet/ddnet/pull/9788 02:23 < bridge> https://tenor.com/view/mercy-overwatch-come-here-gif-14280244 02:24 < bridge> we're getting tune lock before silksong i cant believe it 02:24 < bridge> https://tenor.com/view/gigachad-chad-gif-20773266 02:27 < bridge> it should also supercede #5732 02:27 < bridge> https://github.com/ddnet/ddnet/pull/5732 02:36 < bridge> :D 02:36 < bridge> I didn't see this before 02:39 < bridge> thats.. old 02:41 < bridge> I've often tried to push things *people* wanted but devs/admins didn't see/wanted to, so yeah might not be best implementation and not 100% finished, but on my server it works really well since that PR is out 02:50 < bridge> we live in a society 03:30 < bridge> wait what was the decided behavior for tune lock overrides? 03:30 < bridge> i think tune lock should always be applied on top of the current tune zone if there is one 03:31 < bridge> wow hecta dev role should've had it since a while ago xd 03:38 < bridge> no, tune zones should have higher priority, as discussed with @corneum before. It's way more usable 03:43 < bridge> i wanted to be role-less ngl 03:44 < bridge> how so? 03:46 < bridge> oh yea i can kinda see why 03:46 < bridge> so you be tuned but still have tune zones affect you 03:47 < bridge> how would you make a tunezone in which players without a tunelock have a 0 force grenade, but the guy with tunelock has a normal nade 03:47 < bridge> if tune zones have higher priority 03:56 < bridge> they should both be added on top of eachother, but i think tunelock tunes for a specific `specific_tune_setting` should take priority over a tune zone's `specific_tune_setting` 06:01 < bridge> i almost said server tunes but im assuming the server should have normal grenade physics to begin with 06:03 < bridge> server tunes should just be an abstraction for tunezone across the whole map 06:05 < bridge> this is a very tricky situation because: 06:05 < bridge> - if tune lock has priority, entering a tune zone for the same tune doesn't do anything 06:05 < bridge> - if tune zones have priority, both tees (one tuned and the other not) will have the same exact tunings 06:06 < bridge> *both tees inside it 06:10 < bridge> i think tune lock prio makes sense 06:10 < bridge> @corneum what parts were u trying to make where its worse? 11:32 < bridge> I dont remembee💀 11:33 < bridge> bor for gor 11:42 < bridge> 50$ gift https://steamecommunuty.com/s/1052980419 11:54 < bridge> this is cool i like this 11:54 < bridge> Should new tune lock override old one, or only apply when no override? So in the second case have to unlock with the corresponding unlock tile and then tune lock works 12:35 < bridge> ok i think ill rewrite to 2nd so this setup works just for one player 12:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345721139476955167/image.png?ex=67c593f9&is=67c44279&hm=4e2353d59025bbfb7d4b2f49bdb45f6d3999f324ee2ef39cc44c707e34fb7a6c& 13:18 < bridge> This is how slides would look like :brownbear: 13:18 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345732072261681235/slides_test1_2025-03-02_13-14-18.mp4?ex=67c59e28&is=67c44ca8&hm=333cfc2b977832bd9855b2b8148c3be548f86136607477281819e4f2ecf9faa2& 13:22 < bridge> arrows pointing in the opposite dir still triggers tho 😬 13:22 < bridge> will this break old clients? 13:22 < bridge> they wont have prediction for it 13:23 < bridge> aa so itll break their antiping 13:23 < bridge> will we need new entities? 13:23 < bridge> also the local character prediction will not look smooth 13:23 < bridge> @hectavoxel 🙈 13:23 < bridge> speedup arrows are not part of the entities 13:23 < bridge> no this is part of the speedlayer 13:24 < bridge> you enter the angle in the ui, so the slides have 360 degrees of freedom, force and max speed is not needed 13:26 < bridge> oo thats cool 13:27 < bridge> only downside so far, they are skippable and there is nothing I can do against this 13:27 < bridge> only downside so far, they are skippable and there is nothing I can do against this, they are not slopes 13:28 < bridge> I could implement true slopes, but the prediction isn't the only thing I'd break with true slopes :justatest: 13:33 < bridge> `vButtonsContainersSettingsControlsJoystickAbsolute` 13:35 < bridge> Java 13:35 < bridge> im not allowed to use static anymore 13:35 < bridge> 😭 13:35 < bridge> a simpler time 13:40 < bridge> wait does a vector of 0 size elements have no size per element? 13:40 < bridge> `&vec[1] == &vec[2]` 13:40 < bridge> if anyone needs a review, ask me. I am constantly asking for reviews and want to give something back 13:41 < bridge> no its padded to 1 byte or more 13:41 < bridge> doesn't a template vector have a type, shouldn't this be the size of the type? 13:41 < bridge> if you're using a struct, it's usually padded up to 4 bytes if you don't use the right pragmas for it 13:42 < bridge> `std::vector vButtonsContainersNamePlateKeyPresses = {{}, {}, {}, {}};` 13:42 < bridge> `class CButtonContainer{};` 13:42 < bridge> `sizeof(CButtonContainer) != 0` 13:43 < bridge> friend mark 13:43 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345738385372745850/image.png?ex=67c5a409&is=67c45289&hm=47120572e404c59b225e201a38e3c44b4319e574cbb29eddaaf90e29dba565c5& 13:44 < bridge> 👀 13:58 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345741942566289448/image.png?ex=67c5a759&is=67c455d9&hm=9be65d753658df9f854b49d8a43d6c6d14488aa7bd0475ad163c443bec84a44a& 13:58 < bridge> thanks git very co0ol 14:02 < bridge> btw is there a reason we are not support VS 2022? 14:04 < bridge> waoh this is awesome 14:04 < bridge> best new tile I've ever seen 14:05 < bridge> we do? 14:05 < bridge> I use vs 2022 exclusively 14:05 < bridge> ``` 14:05 < bridge> [cmake] CMake Error at CMakeLists.txt:26 (project): 14:05 < bridge> [cmake] Generator 14:05 < bridge> [cmake] 14:05 < bridge> [cmake] Visual Studio 16 2019 14:05 < bridge> [cmake] 14:05 < bridge> [cmake] could not find any instance of Visual Studio. 14:05 < bridge> [cmake] 14:05 < bridge> [cmake] 14:06 < bridge> [cmake] 14:06 < bridge> ``` 14:06 < bridge> 14:06 < bridge> VS 2022 is installed? 14:06 < bridge> I get this error in VS-Code 14:06 < bridge> there is some way to pass an agurment to tell it to use vs 2022 but I don't use vscode so idk how 14:07 < bridge> I see the variable, gonna search 14:07 < bridge> ` -G "Visual Studio 16 2019"` 14:09 < bridge> just a configuration issue, when I switch to 2022 I get an error ... 14:09 < bridge> what error 14:10 < bridge> `"EBUSY: resource busy or locked, unlink 'c:\Users\Marvin\Desktop\workspace\ddnet\build\CMakeFiles\CMakeConfigureLog.yaml'".` 14:10 < bridge> `"EBUSY: resource busy or locked, unlink 'c:\Users\User\Desktop\workspace\ddnet\build\CMakeFiles\CMakeConfigureLog.yaml'".` 14:11 < bridge> -\ windows moment 14:11 < bridge> -# windows moment 14:11 < bridge> if you're not using the vs 2022 IDE why not just install vs 2019 for the toolchain 14:11 < bridge> then you won't have issues 14:11 < bridge> or your cmake version is old 14:11 < bridge> because windows doesn't like multiple vs versions 14:12 < bridge> it would have to be pretty old 14:12 < bridge> I have 4 VS versions... 14:12 < bridge> i usually just keep one 14:12 < bridge> `>cmake --version 14:12 < bridge> cmake version 3.31.0 14:12 < bridge> ` 14:12 < bridge> so cmake just auto select the right one 14:12 < bridge> maybe it's too new 14:13 < bridge> Just generate the cmake manually? 14:13 < bridge> maybe just cleanup and manually specify generator 14:13 < bridge> There is a cmake gui 14:13 < bridge> i have 3.30.5 works fine 14:13 < bridge> in vs 2022 you just open cmake project and thats it 14:13 < bridge> although to be fair i just used the vscode cmake extension it just list all the compilers and let me click on one 14:14 < bridge> i haven't installed any actual visual studio for years. i only install vs build tools 14:14 < bridge> I deleted the CmakeFiles directory, now I am just missing ffmpeg apprently 14:15 < bridge> .... 14:15 < bridge> ``` 14:15 < bridge> [cmake] CMake Error at CMakeLists.txt:1887 (file): 14:15 < bridge> [cmake] file COPY cannot find 14:15 < bridge> [cmake] "C:/Users/Marvin/Desktop/workspace/ddnet/ddnet-libs/ffmpeg/windows/lib64/avcodec-61.dll": 14:15 < bridge> [cmake] No error. 14:15 < bridge> ``` 14:15 < bridge> ffmpeg should be included in ddnetlibs 14:15 < bridge> name leak! 14:15 < bridge> oh that's on me 14:15 < bridge> is the folder empty 14:15 < bridge> hi Marvin 14:15 < bridge> yeye I already leaked my proffesional github account, deal with it 14:15 < bridge> hi 14:15 < bridge> love you 14:15 < bridge> I love me too 14:16 < bridge> ok continue the troubleshoot. i'm just here to say hi 14:16 < bridge> just a missing `git submodule update` 14:20 < bridge> wait ddnet is supposed to handle ime? 14:20 < bridge> darn 14:20 < bridge> i mean ye.. 14:20 < bridge> chinese player base was big 14:20 < bridge> an application can be completley blind to ime and work pretty functionaly 14:20 < bridge> its nice to have text editing 14:20 < bridge> pretty sure it is toolkits job 14:21 < bridge> but ddnet stores the candidates, so i pressume they are rendered in ddnet 14:21 < bridge> but we obviously aren't using a UI toolkit 14:21 < bridge> ye 14:21 < bridge> only on windows tho 14:21 < bridge> not anymore B) 14:21 < bridge> in SDL3? 14:21 < bridge> yeah 14:21 < bridge> now its cross platform 14:22 < bridge> so bye bye windows restriction 14:22 < bridge> it's not really a restriction 14:22 < bridge> it's a exclusive fullscreen workaround 14:22 < bridge> and you can should be able to choose between "native" (ibus) and ddnet 14:22 < bridge> oh right that 14:22 < bridge> macos always composite so using a in-engine candidate is actually worse than letting system handle IME 14:23 < bridge> xorg and wayland no longer have the exclusive fullscreen restriction either 14:23 < bridge> not sure about linux. i think compositors always have a way to integrate with ibus or fcitx5 14:23 < bridge> but i think you should be able to choose between ddnet and native 14:23 < bridge> its not like ime candidates are that hard to render 14:23 < bridge> so ye, better just kept it windows exclusive in that case. cuz people can really hate the in-game candidate list 14:23 < bridge> you can already 14:24 < bridge> you can always turn it off 14:24 < bridge> there is a setting to bypass the in-game one. 14:24 < bridge> yes 14:24 < bridge> but i do think default to the ime render player usually see is better 14:24 < bridge> i kinda want to make ddnet also automatically give up on render candidates with windows mode 14:24 < bridge> except on windows 14:25 < bridge> I managed to get it working, I needed to manually delete cmake cache and all older build directories 14:25 < bridge> in exclusive fullscreen 14:25 < bridge> because windows moment 14:25 < bridge> windowed fullscreen on windows is weird 14:25 < bridge> the IME works on some machine but not others 14:25 < bridge> its just a window without decoration at full res? 14:25 < bridge> maybe sometime driver might just decide that it should take over the whole screen anyway 14:26 < bridge> to reduce latency for windowed fullscreen 14:26 < bridge> i'm guessing 14:26 < bridge> well since "exclusive fullscreen" is no different for me anymore 14:26 < bridge> i assume all issues with it are gone 14:26 < bridge> not that i ever had any 14:26 < bridge> debugging the server crashes immediatly ... ........ 14:26 < bridge> segfault? 14:27 < bridge> accidental anti-debugger. ~~we should make that a feature and sell them to commercial project~~ 14:28 < bridge> ``` 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Users\Marvin\Desktop\workspace\ddnet\build\release\DDNet-Server.exe'. Module was built without symbols. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\ntdll.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\kernel32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\KernelBase.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\apphelp.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\shlwapi.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\msvcrt.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\ws2_32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\rpcrt4.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\version.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\ole32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\ucrtbase.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\combase.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\gdi32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\win32u.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\gdi32full.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\msvcp_win.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\user32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\shell32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\advapi32.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\sechost.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\winbuilds\bin\sqlite3.dll'. 14:28 < bridge> DDNet-Server.exe (17248): Loaded 'C:\Windows\System32\bcrypt.dll'. 14:28 < bridge> The program '[17248] DDNet-Server.exe' has exited with code -1073741515 (0xc0000135). 14:28 < bridge> ``` 14:28 < bridge> what why does it use release? 14:28 < bridge> WHAT 14:28 < bridge> 😬 14:28 < bridge> -DDEV=ON maybe? 14:29 < bridge> even then debugger shouldn't crash a release build server tho 14:29 < bridge> DDNet-Server.exe (2852): Loaded 'C:\Users\Marvin\Desktop\workspace\ddnet\build\debug\DDNet-Server.exe'. Symbols loaded. 14:29 < bridge> 14:29 < bridge> Still crashes the exact same way 14:29 < bridge> omg 14:30 < bridge> DeepSeek-R1 could be used for coding teeworlds code. 14:30 < bridge> terminal time: 14:30 < bridge> ``` 14:30 < bridge> 2025-03-02 14:30:17 E server: failed to load map. mapname='dm1' 14:30 < bridge> (base) 14:30 < bridge> ``` 14:30 < bridge> xD 14:31 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1345750250127884371/image.png?ex=67c5af16&is=67c45d96&hm=3900266294c9cc99e8c94948bde6df42287e08debd657bee3b9329a8e65cb687& 14:31 < bridge> I asked it something, and then it even know that needing to modify bam.lua 14:31 < bridge> oh is that one of the 14:31 < bridge> fully automated 14:32 < bridge> probably not devin 14:33 < bridge> However, sadly, it couldn't be used in my project about teeworlds. 14:33 < bridge> Now I got it running 14:46 < bridge> is it known, that IsGrounded is 1 pixel off? 14:47 < bridge> I did: 14:47 < bridge> ``` 14:47 < bridge> if(IsGrounded()) 14:47 < bridge> Vel.y = 0; 14:47 < bridge> ``` 14:47 < bridge> And I am flying 1 pixel in the air 14:54 < bridge> something something gravity or something 14:54 < bridge> :kek: 15:04 < bridge> never hang in a 1tile gap? it even makes the joump sound even tho your feet are in air 😄 15:05 < bridge> ```cpp 15:05 < bridge> #if defined(CONF_PLATFORM_ANDROID) 15:05 < bridge> // Translate the Android back-button to the escape-key so it can be used to open/close the menu, close popups etc. 15:05 < bridge> if(Key == KEY_AC_BACK) 15:05 < bridge> { 15:05 < bridge> Key = KEY_ESCAPE; 15:05 < bridge> } 15:05 < bridge> #endif 15:06 < bridge> ``` 15:06 < bridge> ac back is not android specifoc and *could* be generated by a funky controller or keyboard 15:07 < bridge> but the feature is also only active on android xd 15:10 < bridge> right now it is 15:11 < bridge> but theres no reason not to handle it on other platforms 15:11 < bridge> ah so you _want_ it to be used for controllers? 15:11 < bridge> a *funky* controller could 15:11 < bridge> i dont use controller 15:17 < bridge> 15:17 < bridge> `candidates` doesnt seem to disable native rendering 15:33 < bridge> `SDL_EVENT_TEXT_EDITING_CANDIDATES` waiit no way, did sdl actually implement this? 15:33 < bridge> I looked into this for several weeks seeing if I could implement it for SDL but there just was no way to get the candidate list out on macOS 15:33 < bridge> Nor on linux 15:33 < bridge> I wonder how they managed to do it 15:34 < bridge> all the ime methods probably have thier own ways, it just wasnt implemeneted 15:34 < bridge> but now theres enough to do so 15:34 < bridge> maybe macos is still borked 15:34 < bridge> If they didn't implement it for anything except windows I'm very very disappointed 😄 15:34 < bridge> that could be the case because its not hiding the candidate list 15:34 < bridge> that could be the case because its not hiding the candidate ui 15:35 < bridge> that could be the case because its not hiding the candidate ui for me with ibus 15:37 < bridge> on macOS I managed to extract the candidates directly from memory by disabling many security measures and injecting myself into appkit, definitely not a proper solution 15:37 < bridge> ：sob: 15:37 < bridge> +😭 15:37 < bridge> 😭 15:37 < bridge> on Linux there is actually 0 documentation, none of the used IMEs have any of their own documentation, all of their irc channels are dead too 15:38 < bridge> at that point I was like damn, I'm not wasting more time on this shit 15:40 < bridge> There is some standard for it inside X iirc but it wasn't enough 15:43 < bridge> welp 15:43 < bridge> @learath2 do you happen to have a windows machine which can compile ddnet available to you 15:44 < bridge> sure, why? 15:44 < bridge> can you try my sdl3 branch 15:44 < bridge> and see if ime works 15:44 < bridge> Tomorrow I can 15:44 < bridge> 15:44 < bridge> sure 15:44 < bridge> if it works on windows then its good enough 15:45 < bridge> by ime works i mean "you can use it" and "ddnet renders ime candidates" 15:45 < bridge> Looking at it it seems they actually went through the trouble of interacting with each and every possible IM Server on Linux 15:46 < bridge> Props to whoever had the patience to navigate through that mess and figure out how to handle it for each server 15:46 < bridge> can u see where they get candidate lists? 15:47 < bridge> ```cpp 15:47 < bridge> SDL_DBusContext *dbus = SDL_DBus_GetContext(); 15:47 < bridge> 15:47 < bridge> if (IBus_CheckConnection(dbus)) { 15:47 < bridge> Uint32 caps = IBUS_CAP_FOCUS; 15:47 < bridge> 15:47 < bridge> if (hint && SDL_strstr(hint, "composition")) { 15:47 < bridge> caps |= IBUS_CAP_PREEDIT_TEXT; 15:47 < bridge> } 15:47 < bridge> if (hint && SDL_strstr(hint, "candidates")) { 15:47 < bridge> // FIXME, turn off native candidate rendering 15:47 < bridge> } 15:47 < bridge> 15:47 < bridge> SDL_DBus_CallVoidMethodOnConnection(ibus_conn, ibus_service, input_ctx_path, ibus_input_interface, "SetCapabilities", 15:47 < bridge> DBUS_TYPE_UINT32, &caps, DBUS_TYPE_INVALID); 15:47 < bridge> } 15:47 < bridge> ``` 15:47 < bridge> thats probably why it doesnt work for me 15:48 < bridge> could someone merge ? 15:48 < bridge> Yep, it doesn't turn off native rendering 😄 I was just looking at that exact same location 15:48 < bridge> it would be weird to have non native rendeing at the same time 15:48 < bridge> so it probably turns that off but i havent found that yet 15:52 < bridge> Doesn't seem like they do. Only the windows backend seems to call `SDL_SendEditingTextCandidates` 15:52 < bridge> thats too bad ): 15:52 < bridge> well windows is the only platform that benefits from it 15:52 < bridge> so 15:52 < bridge> ¯\_(ツ)_/¯ 15:53 < bridge> Not surprised. Except for windows this stuff is so sparsely documented that the person implementing it probably got bored 15:53 < bridge> https://ibus.github.io/docs/ibus-1.5/IBusInputContext.html 15:54 < bridge> Uncommon windows W 15:54 < bridge> `The "show-lookup-table" signal` 15:54 < bridge> `The "cursor-down-lookup-table" signal` 15:54 < bridge> `The "cursor-up-lookup-table" signal` 15:54 < bridge> you can get the candidate list i think 15:54 < bridge> i cant find out how to hide native ui 15:55 < bridge> classic 15:55 < bridge> Yeah there seems to be a way, `IBUS_CAP_LOOKUP_TABLE` there is this capability indicating you can render your own, maybe setting this is enough? 15:55 < bridge> having extreme dajavu now i'm seeing lerato joined the convo 15:56 < bridge> the wording is vague 15:56 < bridge> `The panel component does nothing for an application that doesn't support focus.` 15:56 < bridge> i wonder how big sdl3 is 15:56 < bridge> to mess around with 15:56 < bridge> Perhaps you listen to `update-lookup-table` signals for it? 15:57 < bridge> btw candidate list really isn't needed on non windows i think. 15:57 < bridge> but its annoying that its not 15:57 < bridge> why 15:57 < bridge> but its annoying that its not possible to do 15:57 < bridge> and perhaps you emit `hide-lookup-table signal` to hide it? 15:57 < bridge> IME worked fine with a compositor 15:58 < bridge> i agree that its not needed, i just think it could and should be supported regardless 15:58 < bridge> It'd be great if it looked the same everywher 15:58 < bridge> It'd be great if it looked the same everywhere 15:58 < bridge> why? 15:58 < bridge> and if you prefer it to be composited on top, you can get a config for it 15:58 < bridge> people use IME with their own settings 15:58 < bridge> theres already a config for it :D 15:58 < bridge> we don't even have a horizontal candidate list 15:59 < bridge> because 15:59 < bridge> that can be changed 15:59 < bridge> idk I don't like having weird inconsistent experiences between people that we don't control 15:59 < bridge> 😬 but we can't just take away their normal IME 15:59 < bridge> for no reason 15:59 < bridge> What happens if you are playing in exclusive fullscreen and aren't running a compositor? 15:59 < bridge> (xorg and wayland dont have that problem) 16:00 < bridge> (nor does macos) 16:00 < bridge> does linux even support typing with IME with no compositor 16:00 < bridge> (nor does macos) 16:00 < bridge> (its just windows) 16:00 < bridge> Who knows? If we were rendering the candidate list on our own these wouldn't be discussions that need having 16:01 < bridge> at least make it automatic or something 16:01 < bridge> i keep getting DMs about turning off the built-in candidate list 16:01 < bridge> it is annoying 16:01 < bridge> and the people that don't like the way we render it can just disable it and wrestle with their compositing and fullscreen/borderless/desktop fullscreen/whatever fullscreen settings themselves 16:01 < bridge> i wanted to just auto bypass the built-in one if it runs in windowed mode on windows 16:01 < bridge> is it not a setting 16:02 < bridge>