Fixes#12920
VTSLS does not mark snippet completions as such in the initial
completion response - not until we resolve them; however, we do not
touch initial contents of completion during resolution, which led to us
not treating a snippet as such.
Release Notes:
- Fixed snippet completions sometimes being treated as plain text
completions when using VTSLS
Note:
- We have disabled all tests that rely on Postgres in the Linux CI. We
only really need to test these once, and as macOS is our team's primary
platform, we'll only enable them on macOS for local reproduction.
- We have disabled all tests that rely on the font metrics. We
standardized on Zed Mono in many fonts, but our CoreText Text System and
Cosmic Text System proved to be very different in effect. We should
revisit if we decide to standardize our text system across platforms
(e.g. using Harfbuzz everywhere)
- Extended the condition timeout significantly. Our CI machines are slow
enough that this is causing spurious errors in random tests.
Release Notes:
- N/A
---------
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
This fixes a panic in the `SelectionsCollection::first_anchor` when
there was a pending mouse selection and no other selections. Until
recently, this method was only used in vim mode, but as of
53b0720d54,
it's also used in the normal `move_up` and `move_down` actions.
So until recently, the panic that this fixes could only happen in vim
mode.
Release Notes:
- Fixed a crash that could happen when using certain cursor-motion
bindings with a pending mouse selection.
This PR disables the doctests in the `gpui_macros` crate, as they depend
on `gpui` to run.
Since `gpui` depends on `gpui_macros`, we don't really want to add a
dependency on `gpui` (even though it _appears_ to work as a dev
dependency).
Also did some minor stylistic cleanup of some doc comments.
Release Notes:
- N/A
This PR removes the need to use `/rustdoc --index <CRATE_NAME>` and
instead indexes the crates once they are referenced.
As soon as the first `:` is added after the crate name, the indexing
will kick off in the background and update the index as it goes.
Release Notes:
- N/A
This PR updates the rustdoc indexing to be more progressive.
Rather than waiting until we've crawled the entire crate to begin
writing to the database, we instead start writing the docs as we go.
This makes it so you can start getting completions while the indexing is
still running.
Release Notes:
- N/A
This PR updates the spots where we render the fold indicators in editors
to use the `Disclosure` component instead of re-implementing similar UI.
This makes this UI more consistent across Zed.
Release Notes:
- N/A
This ensures that linux panics still contain symbol names. It also
allows us to profile Zed on linux with `perf` and get symbol names.
Release Notes:
- N/A
On most platforms, things were working correctly, but had the wrong
type. On X11, there were some problems with window and display size
calculations.
Release Notes:
- Fixed issues with window positioning on X11
---------
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
Co-authored-by: Mikayla <mikayla@zed.dev>
I found that builds failed on Arch and OpenSUSE so I added missing
dependencies. I also found that OpenSUSE Leap is currently not able to
install the required dependencies so I added a check to limit the
supported distros to Tumbleweed.
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
It was possible to unload a root of worktree when it was a .git
directory; due to that, test_fs_events_in_dot_git_worktree was sometimes
stuck in an infinite loop on CI.
The gist of an issue is that when .git dir is a root dir, then modifying
a file within this directory could sometimes unload the .git dir; the
test went into an infinite loop when the first event in an filesystem
stream was not the event for the file creation, but for a dir
modification. In that case we'd unload the root directory and a
subsequent event for file creation would never be registered, leading to
the test being stuck waiting for it to happen.
This commit alleviates it by special-casing worktrees rooted in .git
directories.
Release Notes:
- Fixed a possible hang when opening a worktree in .git directory.
This PR updates the `/rustdoc` command with persistence for the
documented rustdoc items.
Now when you run `/rustdoc --index <CRATE_NAME>` it will index the crate
and store the results in LMDB.
The documented items will then be read from the database when searching
using `/rustdoc` and persist across restarts of Zed.
Release Notes:
- N/A
This was broken by two things:
1. A merge conflict in the install.sh script leading to bad sh syntax
2. A return removed by accident when we refactored main
Release Notes:
- N/A
When working on a repro for a different issue that involved a worktree
with lots of files (100k to be precise), UI became pretty unresponsive.
I pinned it down to us repeatedly preparing a HashSet of all paths in
the currently-scrolled-to worktree, once per each entry in the range
passed to for_each_visible_range (which is e.g. called during
rendering).
This PR makes that hashing happen just once per worktree. Additionally,
we no longer iterate over (potentially) all entries in a given worktree
when calculating the depth of a given entry.
Note that we could probably be smarter about this still; instead of
recalculating the hashset per each call to for_each_visible_entry, we
could do it whenever we update entries in the project panel. However,
with this PR I wanted to get a quick bang for a small buck; I'm pretty
confident in the change as is, it is relatively straightforward and
messing with worktree updates is more involved.
Release Notes:
- Improvement performance of project panel in large worktrees
> * There are a couple of other `zed` binaries that may be present on
linux systems
([1](https://openzfs.github.io/openzfs-docs/man/v2.2/8/zed.8.html),
[2](https://zed.brimdata.io/docs/commands/zed)). If you want to rename
our CLI
binary because of these issues, we suggest `zedit`, `zeditor`, or
`zed-cli`.
Due to aformentioned issue don't hardcode the executable name in the
.desktop file so envsubst can change it in accordance with the
distributor's requirement.
Resolves#12290.
Release Notes:
- N/A
