Time  Nick      Message
19:02 Krock     hmm. To solve the SDL input issue we could in fact use event.text.text which is fired as SDL_TEXTINPUT right after SDL_KEYDOWN
19:03 Krock     SEvent::SKeyInput::Char would then be filled with that character (if possible) and dispatched in SDL_TEXTINPUT (one event later)
19:05 Krock     for PressedDown=false we could track the pressed scancodes in CIrrDeviceSDL
19:05 Krock     Is this idea stupid?
19:06 [MatrxMT] <grorp> we've already discussed this workaround a few times. the reason it doesn't work is: the sdl textinput event is only fired if sdl text input was started before. this will enable ime, which you definitely don't want in-game.
19:19 [MatrxMT] <y5nw> Also: for (particularly) some Asian-language IMEs there is basically no guarantee of keypress events firing at all.
19:42 Krock     ah I see. Thanks IME, I guess.
19:53 jonadab   IMEs can be very important for the kinds of applications where you need to enter actual text, especially if you ever need to enter text in a writing system other than the Latin alphabet.  But yeah, they are completely inappropriate for other types of keyboard-control UIs.
19:59 [MatrxMT] <y5nw> Aside from IMEs, using text input events would complicate key comparisons further. The current KeyPress::operator== checks whether the EKEY_CODE or the key character match (one is enough); this would no longer work since 7 and Shift-7 (let's say it's "/") would need to be considered different keys even if they get the same keycode, while A and Shift-A (moving forward while sneaking) would likely need to be considered the same even with
20:02 celeron55 i think it's simply a bad premise to attempt to use keys that require key combinations in some languages as mapped keys, or even keys that map to multiple possible keys. but doing so was the default for a long time so it has to be unlearned by users
20:05 [MatrxMT] <y5nw> One thing we could do would be to followup the scancode PR with a PR that allows modifier keys; the latter would be a bit weird (since e.g. Shift-7 would be a combination of scancodes and keycodes) but it would allows players to continue using Shift-7 if they wish.
20:09 [MatrxMT] <y5nw> That said, with #14964 we will still end up (one way or another) requiring requiring some users to rebind their keys regardless of what we choose as the default. Scancode-based defaults basically break keyboard layouts where a key is located in a different position than its US counterpart, while keycode-based configurations would break on layouts where a key is not present in the unshifted position.
20:09 ShadowBot https://github.com/minetest/minetest/issues/14964 -- SDL: Use scancodes for keybindings by y5nw
20:18 celeron55 i agree supporting modifiers would probably make enough people happy
20:27 [MatrxMT] <y5nw> I also have a PR for keybindings with modifiers (#14874) but it needs to be reworked to handle scancodes.
20:27 ShadowBot https://github.com/minetest/minetest/issues/14874 -- Allow keybindings with modifiers by y5nw