Fixes https://github.com/zed-industries/zed/issues/7799 by forcing the
modal to close before dispatching the action.
While not needed specifically for this case, changed the context menus
to do the same, to be uniform — context menu actions seem to work
properly after this change too.
Release Notes:
- Fixed markdown preview action not working
([7799](https://github.com/zed-industries/zed/issues/7799))
In Ruby `_$=@!:?` can all be part of an identifier. If we don't have
them in this list, autocomplete doesn't work as expected.
See: https://gist.github.com/misfo/1072693
This fixes one part of #7819 but not the whole ticket.
Release Notes:
- Fixed completions in Ruby not working for identifiers that start or
end with special characters (e.g.: `@`)
This should eliminate a pretty significant (multiple seconds) slowdown
that new users (or users after restarting their OS) have been
experiencing.
Previously, we would just notarize the application, which meant that
every user of the application had to perform an integrity check against
Apple's servers to ensure the app wasn't malicious.
With this commit, we are now using `xcrun stapler staple`, which
attaches the notarization ticket to both the app bundle as well as the
DMG. This should prevent users from needing to reach out to Apple's
notarization service in order to verify the app's integrity.
You can confirm the quarantine status of the application by running `ls
-l@` in `Terminal.app`:
ls -l@ /Applications/Zed.app/Contents/MacOS/zed
Release Notes:
- Improved startup time when opening Zed for the first time or after
restarting the operating system.
Co-authored-by: Thorsten <thorsten@zed.dev>
Co-authored-by: bennetbo <bennetbo@gmx.de>
Co-authored-by: Martin Palma <m@palma.bz>
Co-authored-by: evrsen <146845123+evrsen@users.noreply.github.com>
This PR adds Wayland support to gpui using
[wayland-rs](https://github.com/Smithay/wayland-rs). It is based on
[#7598](https://github.com/zed-industries/zed/pull/7598).
It detects Wayland support at runtime by checking the existence of the
`WAYLAND_DISPLAY` environment variable. If it does not exist or is
empty, the X11 backend will be used. To use the X11 backend in a Wayland
session (for development purposes), you just need to unset
WAYLAND_DISPLAY (`WAYLAND_DISPLAY= cargo run ...`).
At the moment it only creates the window and renders the initial content
provided by `BladeRenderer`, so it can run "Hello world" example.
Todo:
- [x] Add basic Wayland support.
- [x] Add window resizing.
- [x] Add window closing.
- [x] Add window updating.
- [ ] Implement input handling, fractional scaling, and support other
Wayland protocols.
- [ ] Implement all unimplemented todo!(linux).
- [ ] Add window decorations or use custom decorations (like on MacOS).
- [ ] Address other missing functionality.
Release Notes:
- N/A
---------
Co-authored-by: gabydd <gabydinnerdavid@gmail.com>
Co-authored-by: Mikayla Maki <mikayla@zed.dev>
Making media up for release notes / tweets is becoming very time
consuming. I spoke to Max and suggested we ask users to submit media for
their features, to reduce what we need to produce for tweets and such. I
dont know if this is the best way to signal it; I don't like adding more
to the PR template, but I'm not sure of a better way at the moment.
Release Notes:
- N/a
Alternative to #7758, which doesn't involve adding a new trait method
`request_draw`.
Somehow, my whole screen goes blinking black with this when moving the
window, so not ready for landing.
Release Notes:
- N/A
---------
Co-authored-by: Mikayla Maki <mikayla@zed.dev>
This fixes#7523 by enabling completions with placeholders by default.
This setting controls whether gopls sends back snippets with
placeholders. According to the documentation
(https://github.com/golang/tools/blob/master/gopls/doc/settings.md#useplaceholders-bool)
this only controls whether "placeholders for function parameters or
struct fields" are sent in completion responses.
In practice, though, this seems to also control whether any snippets
with *any* placeholders are being sent back.
Example: for the given Go code
err := myFunction()
i^
With the cursor being at `^`, this setting controls whether `gopls`
sends back statement snippets such as `if err != nil { return ... }`
with the `...` being dynamically matched to the return value of the
function.
So I think this setting controls far more than just function params and
struct fields. And since we *do* support placeholders in snippets, I
think this provides a better default experience.
Release Notes:
- Improved default Go experience by enabling snippets-with-placeholders
when initializing `gopls`.
([#7523](https://github.com/zed-industries/zed/issues/7523)).
This PR gives the items in the extension list an explicit height so that
they work properly within the uniform list when descriptions are
missing.
<img width="1235" alt="Screenshot 2024-02-14 at 10 19 14 AM"
src="https://github.com/zed-industries/zed/assets/1486634/01222902-6b05-4e9a-bb5a-bada14b1fd45">
I think we may want to consider using a `list` here instead of a
`uniform_list` to allow them to have variable heights.
Fixes#7756.
Release Notes:
- N/A
This fixes#5211 and #7732 by fixing the case-only file renaming.
The fix here works by checking hooking into function that produces the
data to populate the project panel.
It checks whether we're on a case-insensitive file system (default on
macOS, but you can have case-sensitive FS on macOS too) and if so, it
ignores the metadata for files for which the absolute path (returned by
the FS scanner) and canonicalized path do NOT match.
That's the case for (a) symlinks and (b) case-only renames of files.
It only does this check for case-only renames.
Release Notes:
- Fixed case-only renaming of files producing duplicate entries in
project panel.
([#5211](https://github.com/zed-industries/zed/issues/5211)).
Co-authored-by: Antonio <antonio@zed.dev>
I added this when porting vim mode to gpui2 to work around life-cycle
problems.
Since #7647, this is no longer needed for vim mode, and causes other
problems (c.f. #7748)
Release Notes:
- Improved command to drop fewer keystrokes
This PR adds a view for installing extensions within Zed.
My subtasks:
- [X] Page Extensions and assign in App Menu
- [X] List extensions
- [X] Button to Install/Uninstall
- [x] Search Input to search in extensions registry API
- [x] Get Extensions from API
- [x] Action install to download extension and copy in /extensions
folder
- [x] Action uninstall to remove from /extensions folder
- [x] Filtering
- [x] Better UI Design
Open to collab!
Release Notes:
- Added an extension installation view. Open it using the `zed:
extensions` action in the command palette
([#7096](https://github.com/zed-industries/zed/issues/7096)).
---------
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Carlos <foxkdev@gmail.com>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
Co-authored-by: Max <max@zed.dev>
This does two things:
1. It optimizes the constructions of `SumTree`s to not insert nodes
one-by-one, but instead inserts them level-by-level. That makes it more
efficient to construct large `SumTree`s.
2. It adds a `from_par_iter` constructor that parallelizes the
construction of `SumTree`s.
In combination, **loading a 500MB plain text file went from from
~18seconds down to ~2seconds**.
Disclaimer: I didn't write any of this code, lol! It's all @as-cii and
@nathansobo.
Release Notes:
- Improved performance when opening very large files.
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Julia <julia@zed.dev>
At some point go.mod and go.work syntax highlighting quit working. Looks
like the grammars weren't matching for some reason and I'm not sure how
they were working originally.
Not sure if we could write a test to make sure the tree-sitter queries
are being loaded for the grammars or not but seems like something that
could be useful to avoid something like this in the future.
We see some panics in the Drop handler for MacWindow
Looking into this, I noticed that our drop implementation was not
correctly
cleaning up the window state.
Release Notes:
- N/A
Fixes#4712, or at least works around it. As discussed in the issue,
setting the buttons frames while fullscreen can cause them to render
in the wrong location. This fixes that by simply not moving them when
fullscreen.
Release Notes:
- Separated Linux platform and X11-specific code, so that we can add
Wayland support now.
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
