when no result found, always display `Search all files`, which is
confused.
Release Notes:
- Fixed an issue where the project search would sometimes show "Search
all files" when there were no results.
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This does two things.
Important one: it reverts #19193, which lead to our whole process
handling breaking. When the `proxy` process was killed, it apparently
didn't close the stdout/stderr anymore, which meant we would not detect
when it died. (Watching its `status()` in the io loop also didn't work!)
We should figure out how to keep our process handling working before we
make this change in #19193, which sounds reasonable.
Second, less important thing: I think we should treat the process being
killed from a signal as non-zero, as an error.
Release Notes:
- N/A
This improves the heartbeat detection logic. We now treat any other
incoming message from the ssh remote server
as a heartbeat message, meaning that we can detect re-connects earlier.
It also changes the connection handling to await futures detached.
Co-Authored-by: Thorsten <thorsten@zed.dev>
Release Notes:
- N/A
---------
Co-authored-by: Thorsten <thorsten@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
TL;DR: Another O(n^2) strikes.
In #19194 we received a report about a 7Mb JSON file that Zed struggles
with. Naturally this file showcased a O(n^2) in line layout; this file
has one long line.
During line layout for Mac we have to convert between UTF-16 and UTF-8
indices in the string, as CoreText works with UTF-16 and Rust strings
are UTF-8. The problem stemmed from the fact that we were re-seeking our
string converter on each glyph, which boils down to: we were reparsing
[0..curr_string_position] bytes up to full length of the string, which
is the O(n^2) in question. This PR changes this behaviour to reuse the
Index Converter if the position we're seeking to is not yet reached.
Basically, we're treating the converter as forward iterator and we try
to seek with the same iterator, if possible.
Where previously you could not even open the file in OP (within
reasonable time frame, I waited for 40 seconds before giving up), now
you can do it in.. slightly over a second. The best part is: the
experience is still not ideal. Typing in the buffer is sluggish. Still,
this is a start.
Release Notes:
- Mac: Improved performance with very long lines
This PR changes the SSH modal design so its more keyboard
navigation-friendly and adds the server nickname feature.
Release Notes:
- N/A
---------
Co-authored-by: Danilo <danilo@zed.dev>
Co-authored-by: Danilo Leal <67129314+danilo-leal@users.noreply.github.com>
This also puts the entire state of `StripeBilling` behind a `RwLock`.
When fetching the existing prices and meters, or when inserting new
ones, we acquire a write lock and hold it until the Stripe request
completes. This prevents two concurrent calls to `get_or_insert_price`
from inserting the same data twice.
Creating a new meter/price is unusual, so in practice we'll acquire a
read lock most of the time.
/cc @rtfeldman @maxdeviant
Release Notes:
- N/A
Closes#19051Closes#19182
#### How to reproduce this crash:
1. Open any file and input some ASCII characters.
2. Replace these characters with `你好`.
3. Press `backspace`.
4. Crash.
https://github.com/user-attachments/assets/ea5c5340-29a5-42c8-98c5-6e60770445a4
The issue lies with the `prefix_offset` introduced in #18858. After the
buffer is modified, this value is not always valid and may fall within a
`char boundary`, which results in a crash.
Release Notes:
- Fixed Supermaven crashing on deleting non-ASCII text
remote: Disable ControlPersist for master ssh connection
`ControlPersist=yes` combined with `ControlMaster=yes` silently forces
`ForkAfterAuthentication=yes` (even when the user has explicitly set it
to `no` - reported upstream in [0]) - and the latter makes the ssh
subprocess disappear, which makes us think that the connection died
(This is only an issue for people who have `ControlPersist=yes` in their
`ssh_config`, and perhaps the answer is "if that option breaks things,
don't use that option?" - but it's an option that makes sense _most_ of
the time, it's just in this edge-case of "creating an ssh connection
with -N and expecting the process to stay in the foreground" where it
_must_ be set to no)
I think the alternative approach is to tell people "if you want to use
persistent connections, have a separate ~/.ssh/config entry for
servername (to ssh into) and servername-no-persist (to zed into)", which
is possible, but ugh. Kind of a messy situation >.<
Tests:
- Before: Connections to my server result in "Failed to connect: ." (The
error message is attempting to show stderr, but stderr is empty)
- After: Connections to my server work reliably
[0] https://bugzilla.mindrot.org/show_bug.cgi?id=3743
Release Notes:
- N/A
- Closes#16998
This PR resolves issues with the /file and /diagnostics commands in the
assistant panel, which previously failed to display the contents of a
directory when searching for a folder instead of using the arrow button.
- Changed the format in `project.rs` (located at
`crates/project/src/project.rs`) to use `std::path::MAIN_SEPARATOR` for
cross-platform compatibility, which resolves errors encountered on
Windows that originally used the format `format!("{}/", ...)`.
Release Notes:
- N/A
Similar to e2647025ac, this adds a `regex`
option to `vim::Search` command to allow disabling regex search.
Release Notes:
- Added `regex` option to `vim::Search` command to allow disabling regex
search by default in the keymap. Example usage:
```yaml
{
"context": "VimControl && !menu",
"bindings": {
"/": ["vim::Search", { "regex": false }],
}
}
```
This will allow us to compile debug builds of the remote-server for a
different architecture than the one we are developing on.
This also adds a CI step for building our remote server with minimal
dependencies.
Release Notes:
- N/A
Closes#19131Closes#19039
fixes the broken auto-updater.
I had the bright idea of using streams as the most common unit of data
transfer. Unfortunately, streams are not re-usable. So HTTP redirects
that have a stream body (like our remote server and auto update
downloads), don't redirect, as they can't reuse the stream. This PR
fixes the problem and simplifies the AsyncBody implementation now that
we're not using Isahc.
Release Notes:
- N/A
This PR adjusts the billing logic to not write any records to
`billing_events` if:
- The user is staff, as we don't want to bill staff members
- Billing is disabled (we currently enable billing based on the presence
of the Stripe API key)
Release Notes:
- N/A
This PR makes the `StripeBilling` object long-lived so that we can make
better use of the cached data on it.
We now hold it on the `AppState` and spawn a background task to
initialize the cache on startup.
Release Notes:
- N/A
Co-authored-by: Richard <richard@zed.dev>
When setting a remote ssh project path starting with ~, Zed would fail
to cd into such project's directory when opening a new terminal.
Release Notes:
- N/A
---------
Co-authored-by: Thorsten Ball <mrnugget@gmail.com>
This PR adds usage-based billing for LLM interactions in the Assistant.
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Richard <richard@zed.dev>
Co-authored-by: Richard Feldman <oss@rtfeldman.com>
collab: Remove dependency on X11
I'm not sure if this is the best solution (perhaps pulling
`LanguageName` into a separate `language_types` crate would be
better...?) - but it massively reduces build time / dependencies / size
and means that the collab server no longer requires X11 libraries to be
installed.
tl;dr: `telemetry_events` requires the `language` crate, and the
language crate requires a whole ton of extra stuff. Since
telemetry_events only uses `language` for a single type definition
(`LanguageName`, aka `String`), we can cut all of these out by using the
base `String` type (This doesn't seem too terrible, given that all other
telemetry fields are using basic datatypes like String as opposed to
more strongly-typed variants).
FYI the dependency tree for "why does collab need X11 libraries??" looks
like this:
```
collab
\- telemetry_events
\- language
|- gpui
|- fuzzy
| \- gpui
|- git
| \- gpui
|- lsp
| |- gpui
| \- release_channel
| \- gpui
|- settings
| |- fs
| | \- gpui
| \- gpui
|- task
| \- gpui
\- theme
\- gpui
```
Release Notes:
- N/A
This PR adds support to the assistant for display billing-related
errors.
Pulling this out of #19081 to make it easier to cherry-pick.
Release Notes:
- N/A
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Richard <richard@zed.dev>
terminal: Improve default locale handling
* Use `LANG` instead of `LC_ALL` (`LC_ALL` is the highest priority which
will override any other end-user settings; when that isn't set things
fall back to separate `LC_*` variables; and when those aren't set things
fall back to `LANG`). [0]
* Only set `LANG` for our child if necessary (if it already exists in
the parent, then the child will inherit that, no need for us to do
anything)
[0]
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html#tag_08_02
Tested cases:
- `unset LANG ; cargo run`: locale inside zed's terminal is set to
`en_US.UTF-8`
- `export LANG=en_GB.UTF-8 ; cargo run`: locale inside zed's terminal is
set to `en_GB.UTF-8`
Release Notes:
- Use the system locale in the terminal instead of forcing `en_US.UTF-8`
This fixes two things:
- Go-to-def to absolute paths (i.e. opening stdlib files) multiple times
(opening, dropping, and re-opening worktrees)
- Re-opening abs paths from the file picker history that were added
there by go-to-def
Release Notes:
- N/A
---------
Co-authored-by: Bennet <bennet@zed.dev>
Currently terminal.cursor_shape uses `underline` and `cursor_shape` uses
`underscore`.
This standardizes them so they use the same settings value.
I think `underline` is the more common term and it matches the
terminology used by VSCode, Alacritty, iTerm, etc.
Note the protobuf enum `CursorShape::CursorUnderscore` remains
unchanged.
See also:
- https://github.com/zed-industries/zed/pull/18530
- https://github.com/zed-industries/zed/pull/17572
Release Notes:
- Settings: rename one `cursor_shape` from `underscore` to `underline`
(breaking change).
This is a bit of a personal thing, but it's been bugging me for a while
now that the search options are sticky whenever I use `/` in Vim mode.
This change makes it so that the options are reset with each new `/`.
That means you can, for example, use `v` to create a visual selection,
then hit `*` to search for that (which activates a bunch of search
options), but then continue with `/` to get a normal search.
Release Notes:
- Changed `/` in Vim mode to always reset the search options in the
search bar back to regex-only. That means using `*` (in normal or visual
mode) still works with its options, but the next `/` will reset the
search options. That makes it much closer to how `/` behaves in Vim.
- The App Shortcuts in macOS System Settings does not work for Zed since the menu items titles were not set.
- Previously you could set a shortcut for `Zoom`.
- This add support for `Window->Zoom` as well.
As Zed instantly shows completion items in the completion menu, and the
resolve will cause the details to appear, flickering.
We can safely resolve the `documentation`, `additionalTextEdits` and
`command` fields, the rest should be resolved eagerly for now.
Release Notes:
- Fixed completion menu rendering
Track the cursor offset before biasing in the Supermaven completion
provider to better determine if the text should be suggested. The
underlying issue here is due to the way anchor biasing works, the
completion provider is not able to determine if a given suggestion's
cursor location no longer exists as it is always coalesced to a correct
location (specifically, the end of the line).
This change updates that logic so the offset is stored independently of
the buffer so it can be used to represent a location that may not exist
in the buffer anymore to represent locations that have been deleted.
The net effect is that suggestions can be backspaced much more cleanly
with Supermaven.
Release Notes:
- Improves https://github.com/zed-industries/zed/issues/17981 to prevent
suggesting completions based on out-of-date cursor locations.
- Closes#18610
This PR addresses the same issue as PR #18578. After a full day of
research and testing, I believe I’ve found the best solution to resolve
this issue. With this PR, the window creation behavior on Windows
becomes more consistent with macOS:
- When `params.show` is `true`: The window is created and immediately
displayed.
- When `params.show` is `false`: The window is created but remains
hidden until the first call to `activate_window`.
As I mentioned in #18578, `winit` creates hidden windows by setting the
window's `exstyle` to `WS_EX_NOACTIVATE | WS_EX_TRANSPARENT |
WS_EX_LAYERED | WS_EX_TOOLWINDOW`, which is different from the method
used in this PR. Here, the window is created with normal parameters, but
we do not call `ShowWindow` so the window is not shown.
I'm not sure why `winit` doesn't use a smilliar approach like this PR to
create hidden windows. My guess is that `winit` is creating this hidden
window to function as a "DispatchWindow" — serving a purpose similar to
`WindowsPlatform` in `zed`. To ensure the window stays hidden even if
`ShowWindow` is called, they use the `exstyle` approach.
With the method used in this PR, my initial tests haven't revealed any
issues.
Release Notes:
- N/A
