Web: return MonitorHandle in Window::fullscreen()
#3861
Conversation
- Change `OrientationData::natural` type from `bool` to `Orientation`, just telling the user what the natural orientation is. - Fix and improve some monitor related documentation.
daxpedda
force-pushed
the
web-monitor-3
branch
from
be60876 to
2f4144f
Compare
| @@ -312,7 +312,7 @@ impl Inner { | |||
| #[inline] | |||
| pub(crate) fn fullscreen(&self) -> Option<Fullscreen> { | |||
| if self.canvas.is_fullscreen() { | |||
| Some(Fullscreen::Borderless(None)) | |||
| Some(Fullscreen::Borderless(Some(self.monitor.current_monitor()))) | |||
There was a problem hiding this comment.
Not sure if this would necessarily be better? Isn't the "correct" thing to do here to track which state the user originally requested to fullscreen in (custom monitor chosen, or no monitor chosen), and then return the same here? (Such that the assertion in the sequence window.set_fullscreen(fullscreen); assert_eq!(window.fullscreen(), fullscreen) would hold).
There was a problem hiding this comment.
I was unsure as well how this would play out, but after some testing I've learned that the current screen is always the one requested to be fullscreen, regardless of what it was before.
So yes, your assertion would hold.
There was a problem hiding this comment.
I mean, if I set let fullscreen = Fullscreen::Borderless(None), the assertion no longer holds.
There was a problem hiding this comment.
Ah, I see!
No, in that case it won't.
I'm not entirely sure what we should do here exactly, but I'm actually in favor of the current way Web works: always return the currently correct values, regardless of user input.
E.g. what do I do if set_fullscreen() fails? Should the assert hold until I receive the error?
This is all less then ideal and should be documented thoroughly ... so I'm gonna leave a comment in #3690.
MonitorHandleinWindow::fullscreen()OrientationData::naturaltype frombooltoOrientation, just telling the user what the natural orientation is instead of letting the user "do the math".Follow-up to #3801.