We've had several outages with a proximate cause of "vercel is
complicated",
and auto-update is considered a critical feature; so lets not use vercel
for
that.
Release Notes:
- Auto Updates (and remote server binaries) are now downloaded via
https://cloud.zed.dev instead of https://zed.dev. As before, these URLs
redirect to the GitHub release for actual downloads.
- Enable Alt+Numpad input
- For this to be effective, the default keybindings for Alt+{Number}
will need to be unbound. This won't be needed once we gain the ability
to differentiate numpad digit keys from alphanumeric digit keys.
- Fixes https://github.com/zed-industries/zed/issues/40699
- Fix a number of edge cases with dead keys
Release Notes:
- N/A
Technically these should not be possible to hit, but sentry says
otherwise. Turning these into errors should give us more information
than the abort due to unwinding across ffi boundaries.
Fixes ZED-321
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Release Notes:
- N/A
---
https://github.com/user-attachments/assets/d46b77ae-88ba-43da-93ad-3656a7fecaf9
The system menu is only support for macOS, so here just modify the macOS
platform special code.
The Windows, Linux used `ApplicationMenu`, I have already added
`checked` option to Zed's ContextMenu.
Then later when this PR merged, we can improve "View" menu to show check
state to panels (Project Panel, Outline Panel, ...).
Given that when we lose our devices unrecoverably we will panic anyways,
might as well do so eagerly which makes it clearer.
Additionally this PR replaces all uses of `ManuallyDrop` with `Option`,
as otherwise we need to do manual bookkeeping of what is and isn't
initialized when we try to recover devices as we can bail out halfway
while recovering. In other words, the code prior to this was fairly
unsound due to freely using `ManuallyDrop::drop`.
Fixes ZED-1SS
Release Notes:
- N/A *or* Added/Fixed/Improved ...
- On macOS, pasting now inserts the actual file path when the clipboard
contains a file URL (public.file-url/public.url)
- Terminal paste remains text-only; no temp files or data URLs are
created. If only raw image bytes exist on the clipboard, paste is a
no-op.
- Scope: macOS only; no dependency changes.
- Added a test (test_file_url_converts_to_path) that verifies URL→path
conversion using a unique pasteboard.
Release Notes:
- Improved pasting on macOS: now inserts the actual file path when the
clipboard contains a file URL (enables image paste support for Claude
Code)
On macOS, the Function key is reserved for system use and should not be
used in application code.
This commit updated keystroke matching and key event handling to ignore
the Function key modifier while users are typing or pressing
keybindings.
For some keyboards with compact layout (like my 65% keyboard), there is
no separated backtick key. Esc and it shares the same physical key. To
input backtick, users may press `Fn-Esc`. However, macOS will still
deliver events with Fn key modifier to applications. Cocoa framework can
handle this correctly, which typically ignore the Fn directly. GPUI
should also follow the same rule, otherwise, the backtick key on those
keyboards won't work.
Release Notes:
- Fixed a bug where typing fn-\` on macOS would not insert a `.
The Alt+Numpad buffer that's maintained by the input stack is getting
corrupted, leading to garbage characters being inserted on keystrokes
like Alt+Up. Disable the automatic handling of Alt+Numpad for now until
the cause of this corruption is understood. The Alt+Numpad input did not
work anyway, so this does not regress anything.
Release Notes:
- windows: Fixed corrupted characters being inserted when Alt is pressed
(preview only)
Works around an issue, which can be reproduced in the following program:
```rs
use windows::Win32::UI::Input::KeyboardAndMouse::{GetKeyboardState, VK_CONTROL};
fn main() {
let mut keyboard_state = [0u8; 256];
unsafe {
GetKeyboardState(&mut keyboard_state).unwrap();
}
let ctrl_down = (keyboard_state[VK_CONTROL.0 as usize] & 0x80) != 0;
println!("Is Ctrl down: {ctrl_down}");
}
```
In debug mode, this program prints the correct answer. In release mode,
it always prints false. The optimizer appears to think that
`keyboard_state` isn't mutated and remains zeroed, and folds the
`modifier_down` comparisons to `false`.
Release Notes:
- N/A
`ICustomDestinationList::AppendCategory` rejects an empty `IObjectArray`
and returns an `E_INVALIDARG` error. Error propagated and caused an
early-return from `update_jump_list()`.
<img width="1628" height="540" alt="image"
src="https://github.com/user-attachments/assets/f8143297-c71e-42a1-a505-66cd77dfa599"
/>
Release Notes:
- N/A
This reintroduces `layer_shell` support after #32651 was reverted. On
top of that, it allows setting options for the created surface,
restricts the enum variant to the `wayland` feature, and adds an example
that renders a clock widget using the protocol.
I've renamed the `WindowKind` variant to `LayerShell` from `Overlay`,
since the protocol can also be used to render wallpapers and such, which
doesn't really fit with the word.
Things I'm still unsure of:
- We need to get the layer options types to the user somehow, but
nothing from the `platform::linux` crate was exported, I'm assuming
intentionally. I've kept the types inside the module (instead of doing
`pub use layer_shell::*` to not pollute the global namespace with
generic words like `Anchor` or `Layer` Let me know if you want to do
this differently.
- I've added the options to the `WindowKind` variant. That's the only
clean way I see to supply them when the window is created. This makes
the kind no longer implement `Copy`.
- The options don't have setter methods yet and can only be defined on
window creation. We'd have to make fallible functions for setting them,
which only work if the underlying surface is a `layer_shell` surface.
That feels un-rust-y.
CC @zeroeightysix
Thanks to @wuliuqii, whose layer-shell implementation I've also looked
at while putting this together.
Release Notes:
- Add support for the `layer_shell` protocol on wayland
---------
Co-authored-by: Ridan Vandenbergh <ridanvandenbergh@gmail.com>
### Problem
On Windows, the right Alt key was not working in keybindings (e.g.,
`Ctrl+Right Alt+B`), while the left Alt key worked correctly. This was
due to overly aggressive AltGr detection that treated any `right Alt +
left Ctrl` combination as AltGr, even on US keyboards where AltGr
doesn't exist.
### Root Cause
Windows internally represents AltGr (Alt Graph) as `right Alt + left
Ctrl` pressed simultaneously. The previous implementation always
excluded this combination from being treated as regular modifier keys to
support international keyboards. However, this broke keybindings using
right Alt on US/UK keyboards where users expect right Alt to behave
identically to left Alt.
### Solution
Implemented keyboard layout-aware AltGr detection:
1. Added `uses_altgr()` method to `WindowsKeyboardLayout` that checks if
the current keyboard layout is known to use AltGr (German, French,
Spanish, Polish, etc.)
2. Modified `current_modifiers()` to only apply AltGr special handling
when the keyboard layout actually uses it
3. Added explicit checking for both `VK_LMENU` and `VK_RMENU` instead of
relying solely on the generic `VK_MENU`
### Behavior
- **US/UK keyboards**: Right Alt now works identically to left Alt in
keybindings. `Ctrl+Right Alt+B` triggers the same action as `Ctrl+Left
Alt+B`
- **International keyboards** (German, French, Spanish, etc.): AltGr
continues to work correctly for typing special characters and doesn't
trigger keybindings
- **All keyboards**: Both Alt keys are detected symmetrically, matching
the behavior of left/right Windows keys
### Testing
Manually tested on Windows with US keyboard layout:
- `Ctrl+Left Alt+B` triggers keybinding
- `Ctrl+Right Alt+B` triggers keybinding
- Both Alt keys work independently in keybindings
Release Notes:
- Fixed Right Alt key not working in keybindings on Windows
This caused issues with #40172, as it made Zed execute and block on tad
few more background tasks. Parker is ~cheap to create, hence we should
be ok to just create it at the time it is needed.
Release Notes:
- N/A
---------
Co-authored-by: Cole Miller <cole@zed.dev>
Closes#40359
We were segfaulting when opening a UTF-8 file starting with a byte order
mark due to a mismatch in our UTF-16 indexing calculations caused by
Core Foundations `replace_str` stripping the BOM internally. This PR
fixes the crash by replacing one of our manual calculations by calling
the Core Foundations API to get the length of a string.
Release Notes:
- Fixed a crash on macOS when opening a file that starts with a UTF-8
byte order mark (BOM).
Co-authored-by: HactarCE <6060305+HactarCE@users.noreply.github.com>
`TaskDialogIndirect` may return `IDCANCEL` when the user quits the
dialog via escape or alt+f4, so we need to account for that.
Fixes ZED-25H
Release Notes:
- Fixed panic when hitting escape in dialogs on windows
This reverts commit f1db1f3a3c.
This seems to have affected the vertical positioning of text that
doesn't contain emojis in a way that was unintended.
Release Notes:
- N/A
Closes#36934
I'm still experiencing bugs with the
`DCompositionWaitForCompositorClock` API. Let's back out the support for
now until the fixes are identified and widely available.
`DwmFlush` does various things that aren't just waiting for VSync, so
it's not ideal, but it's not bad enough that it's worth a bigger
refactor right now.
Release Notes:
- N/A
This applies the same fix as #39886 for Windows.
Previously we were using `GetLineMetrics` to determine the ascent and
descent values for each line. It seems like this has the same behavior
as `GetTypographicBounds` on macOS, which is to return the minimum
ascent and descent for the current state of the `TextLayout` object.
This causes the ascent/descent to be unstable when adding or removing an
emoji because a font fallback is triggered when an emoji is present on
the line.
The issue is fixed by switching to `font.GetMetrics` to get the ascent
and descent, which should always return stable values for the main font,
instead of changing when there's a fallback. This also should support
situations where we have multiple explicit fonts on the same line,
although that probably can't be triggered in Zed right now.
Release Notes:
- windows: Fixed a vertical shift in text layout when inserting or
removing an emoji.
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
This relands https://github.com/zed-industries/zed/pull/37175 as
https://github.com/zed-industries/zed/pull/39886 fixed the jiggling
issue.
Currently when we render text with differing styles adjacently we might
form a ligature between the text, causing the ligature forming
characters to take on one of the two styles. This can especially become
confusing when a ligature is formed between actual text and inlay hints.
Annoyingly, the only ways to prevent this with core text is to either
render each run separately, or to insert a zero-width non-joiner to
force core text to break the ligatures apart, as it otherwise will merge
subsequent font runs of the same fonts.
We currently do layouting on a per line basis and it is unlikely we want
to change that as it would incur a lot of complexity and annoyances to
merge things back into a line, so this goes with the other approach of
inserting ZWNJ characters instead.
Note that neither linux nor windows seem to currently render ligatures,
so this only concerns macOS rendering at the moment.
Closes https://github.com/zed-industries/zed/issues/23194
Release Notes:
- Fixed ligatures forming between real text and inlay hints on macOS
Closes#39263
Release Notes:
- N/A
from
https://github.com/zed-industries/zed/issues/39263#issuecomment-3358220988
>
> > If you replace that code with
> >
> > let adapter: IDXGIAdapter1 = unsafe {
> > dxgi_factory.EnumAdapters(adapter_index)
> > }?.cast()?;
> >
> > does it not select the right GPU?
>
> @reflectronic That does seem to select the active gpu for me, meaning
whichever GPU is currently connected. This is a much simpler solution
than the one I have here
(https://github.com/zed-industries/zed/pull/39264 - updated) and while
I'm sure I could imagine someone wanting to choose their GPU to render
Zed on, that may not be something that the application really needs to
support.
>
> I have a branch with just this as the only change that I can push to
that PR if the simpler solution is preferred.
>
> ```rust
> let adapter: IDXGIAdapter1 = unsafe {
> dxgi_factory.EnumAdapters(adapter_index)?.cast()?
> };
> ```
Closes #ISSUE
This allows new windows like the Rules library or the Settings UI window
to appear floating on window managers like hyprland:
https://github.com/user-attachments/assets/628db7f9-4459-4601-85f1-789923831182
Left is with `WindowKind::Floating` and right is with
`WindowKind::Normal`
Release Notes:
- Added support for floating windows on x11 and wayland
The previous modifier detection treated `AltGr` presses as `Ctrl+Alt`,
which broke entering characters produced by AltGr. For example, on a
Hungarian layout `{` is typed with `AltGr+B`; our code saw that as
`Ctrl+Alt+B` and the keybind took precedence, so the character couldn’t
be entered.
On Windows, AltGr isn’t a first-class modifier. It’s emulated as a
combination of `Right Alt (VK_RMENU)` plus a synthetic `Left Ctrl
(VK_LCONTROL)` press. When users press AltGr, `GetKeyState` reports both
Ctrl and Alt as down, which makes AltGr indistinguishable from a real
`Ctrl+Alt` chord if we only look at aggregate modifier state.
Fix: detect the AltGr pattern by checking `VK_RMENU && VK_LCONTROL`.
When that pattern is present, treat it as text-entry intent and suppress
`control` and `alt` in `current_modifiers()`. This prevents
AltGr-produced characters from colliding with `Ctrl+Alt` keybinds while
keeping other modifiers intact.
Limitation: there is no Windows API to tell whether the active layout
actually has AltGr. As a result, on non-AltGr layouts (e.g. US),
pressing `Right Alt + Left Ctrl` will be interpreted as AltGr and will
not trigger `Ctrl+Alt` keybinds. This is an acceptable trade-off to
ensure AltGr layouts can reliably enter characters; users can still
invoke `Ctrl+Alt` keybinds using `Left Alt` or by choosing bindings that
avoid common AltGr pairs.
I based this on https://github.com/zed-industries/zed/pull/36115 after
trying other different approaches, but this one is a bit more specific.
Does this approach make sense, or is slightly breaking US input in favor
of fixing international input a no-go? I think the benefit - being able
to type certain characters _at all_ - outweighs the shortcomings.
Otherwise, there's a way to detect if the keyboard layout uses AltGr or
not, but it's quite hacky, and involves reading the registry to find the
current layout dll's name, opening that dll, manually declaring struct
layouts that it uses, then parsing out the AltGr flag from a function
call result. I don't think that's worth it, but if needed, I can give
that a shot, let me know.
Release Notes:
- windows: Fixed handling of AltGr to avoid keybinds preventing
character input
Prior we only logged the crate in `log_err`, which is not too helpful.
We now assemble the module path from the file system path.
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Lukas Wirth
Richard Feldman
Conrad Irwin
TryanksandMikayla Maki
John Tur
Jason Lee
Sean Timm
Cyandev
scuzqy
Antal Szabó
Hilmar WiegandandRidan Vandenbergh
Sushant Mishra
Kirill Bulatov
joel
Piotr OsiewiczandCole Miller
Andrew FarkasandCole Miller
0x2CA
Cave Bats Of Ware
Bennet Bo Fenner
Francisco Gonzalez
SunliandJason Lee
Alvaro Parker