Fixes#2516
Those changes mean that when we have one ghostty window in non-native
fullscreen and another ghostty window not in fullscreen switching to not
fullscreen window won't cause appearing menu bar and dock. I think it
looks good:
If we implement detection and make menu bar and dock appear for not
fullscreen window in this case it will cause the fullscreen window to
change its size and will look bad.
Non-native fullscreen works bad with multiple screens in either way.
E.g. switching to a non-native fullscreen window would cause menubar
disappering on another screen or switching to not fullscreen window
would show menu bar over fullscreen window on another screen. I think
nobody would want non-native fullscreen with multiple screens.
Fixes#2900
It's possible for moveFocus to infinite loop if the surface view we're
trying to move focus to NEVER gets attached to a window. This can happen
if the window is destroyed.
I think this issue should be more systemically fixed so it can't happen
but this workaround for now prevents moveFocus from being an infinite
loop source for the time being.
Fixes#2840
Related to #2842
This builds on #2842 by missing a key situation: when native fullscreen
is toggled using the menu bar items it doesn't go through our
`FullscreenStyle` machinery so we don't trigger fullscreen change
events.
This commit makes it so that our FullscreenStyle always listens for
native fullscreen change (even in non-native modes) to fire a fullscreen
did change event. This way we can always rely on the event to be fired
when fullscreen changes no matter what.
Fixes#2848
The proper way to convert a unicode scalar in Swift is to use the
`String` initializer that takes a `UnicodeScalar` as an argument. We
were converting a number to a string before, which is incorrect.
Fixes#2850
In native fullscreen, the titlebar container is no longer part of our
NSWindow and is instead a separate window called
NSToolbarFullScreenWindow. We now search for this window when we are in
native fullscreen.
The prior light/dark mode awareness work works on surface-level APIs. As
a result, configurations used at the app-level (such as split divider
colors, inactive split opacity, etc.) are not aware of the current theme
configurations and default to the "light" theme.
This commit adds APIs to specify app-level color scheme changes. This
changes the configuration for the app and sets the default conditional
state to use that new theme. This latter point makes it so that future
surfaces use the correct theme on load rather than requiring some apprt
event loop ticks. Some users have already reported a short "flicker" to
load the correct theme, so this should help alleviate that.
Previously, we would access the `ghostty.config` object from anywhere.
The issue with this is that memory lifetime access to the underlying
`ghostty_config_t` was messy. It was easy when the apprt owned every
reference but since automatic theme changes were implemented, this isn't
always true anymore.
To fix this, we move to the same pattern we use internally in the core
of ghostty: whenever the config changes, we handle an event, derive our
desired values out of the config (copy them), and then let the caller
free the config if they want to. This way, we can be sure that any
information we need from the config is always owned by us.
Fixes#2695
We had various issues with the pointerVisible property on macOS,
including the pointer not being hidden when it should be. Our only use
case today is mouse hide while typing so
NSCursor.setHiddenUntilMouseMoves is a better fit!
This allows us to enrich the build's commit property as a GitHub link.
This change also displays the property values using a monospaced font,
which I think looks a little nicer (especially the commit SHA).
The bug: ctrl+shift+enter on macOS 15 shows a context menu and doesn't
encode to the terminal.
This avoids a system-wide keybind that shows a context menu in macOS
15+. In general Ghostty doesn't try to override system-wide keybinds
but this one is particularly annoying and not useful to terminal users.
We've discussed making this logic configurable for all system level
keybinds but for now this is a quick fix specifically for
ctrl+shift+enter.
Fixes#2565
This appears to be a bug in macOS 15. Specifically on macOS 15 when the
new native window snapping feature is used, `cascadeTopLeft(from: zero)`
will move the window frame back to its prior unsnapped position.
The docs for `cascadeTopLeft(from:)` explicitly say:
> When NSZeroPoint, the window is not moved, except as needed to constrain
> to the visible screen
This is not the behavior we are seeing on macOS 15. The window is on the
visible screen, we're using NSZeroPoint, and yet the window is still
being moved. This does not happen on macOS 14 (but its hard to say
exactly because macOS 14 didn't have window snapping).
This commit works around the issue by saving the window frame before
calling `cascadeTopLeft(from: zero)` and then restoring it afterwards
if it has changed.
I've also filed a radar with Apple for this issue.
Fixes#2537
This matches Terminal.app. iTerm2 requires cmd+option (our old
behavior). Kitty doesn't seem to support rectangle select or I couldn't
figure out how to make it work. WezTerm matches Terminal.app too.
Outside of terminal emulators, this is also the rectangular select
binding for neovim.
Fixes#2519
This sets up the colorspace for terminal windows in the base controller.
This also modifies some of our logic so its easier for subclasses of
base controllers to specify custom logic when the configuration reloads,
since that's likely to be a common thing.
