Wayland: improve keyboard handling #7970
Conversation
cla-bot
bot
added
the
cla-signed
|
|
||
| // Need to implement state.repeat_delay and state.repeat_rate | ||
| // todo!(linux) | ||
| state.repeat_current_keysym = Some(key_sym); |
There was a problem hiding this comment.
Key repeat must be handled by the wayland client.
I'd do it myself but I don't understand async rust and the executor well enough. IIUC, to schedule a task to run after a timeout I should do something like this? Or should I cancel the task in the key-release event? If so, how do I cancel a task?
state.platform_inner.foreground_executor.spawn(async move {
Timer::after(state.repeat_delay).await;
if let Some(key_sym) = state.repeat_current_keysym {
trigger_keypress_repeat();
}
})
| xkb::Keysym::BackSpace | ||
| | xkb::Keysym::Escape | ||
| | xkb::Keysym::Home | ||
| | xkb::Keysym::End | ||
| | xkb::Keysym::Delete | ||
| | xkb::Keysym::Insert | ||
| | xkb::Keysym::Menu | ||
| | xkb::Keysym::Left | ||
| | xkb::Keysym::Right | ||
| | xkb::Keysym::Down | ||
| | xkb::Keysym::Up | ||
| | xkb::Keysym::Page_Down | ||
| | xkb::Keysym::Page_Up | ||
| | xkb::Keysym::Super_L | ||
| | xkb::Keysym::Super_R |
There was a problem hiding this comment.
tab and space are inputed as \t and \s, should they be in this list? What we really want is a is_keysym_printable() function, but it always ends up as a list.
There's missing keysyms here. We might want to use a list from somewhere that has already done this more thouroughly, e.g. from chromium.
There was a problem hiding this comment.
I'm also looking into translating control characters into the key pressed, so that we could use ctrl-(key) keymaps. (https://www.geeksforgeeks.org/control-characters/) #7975
There was a problem hiding this comment.
Isn't the code below doing the work that you're trying to do there?
See line 407-408 of this file post patch, right below.
There was a problem hiding this comment.
Yep, it does... My bad, and thank you, i'll close my PR
| } else if key_sym == xkb::Keysym::Super_L | ||
| || key_sym == xkb::Keysym::Super_R | ||
| { | ||
| state.modifiers.command = true; |
There was a problem hiding this comment.
For the time being, this allows to test keyboard shortcuts (and features) by using super instead of cmd as there isn't a linux keymap yet.
Not sure what behavior we want, but we could use the same convention as the web: both the super and command keys map to the "meta" modifier.
There was a problem hiding this comment.
It doesn't seem like a good desicion because it doesn't work on all platforms (e.g. Gnome)
|
Found this PR just as I was adding the else if for super to command modifier 😄 . I agree that super, cmd and windows key should map to a meta key. Also, I would add a comment to gpui/src/platform/keystroke.rs, right above the pub struct Modifiers {
....
/// The command key, on macos
/// the windows key, on windows
/// the super key, on linux
pub command: bool,
....
} |
|
I've just opened #7989, which I think might solve the same problems as this issue without having to maintain a list of keycodes here. |
That works for me, avoiding storing the modifiers is a better approach as it will make |
Makes keyboard somewhat more usable under wayland.