This PR replaces the `lazy_static!` usages in the `paths` crate with
`OnceLock` from the standard library.
This allows us to drop the `lazy_static` dependency from this crate.
The paths are now exposed as accessor functions that reference a private
static value.
Release Notes:
- N/A
Previously we were using a single globset::Glob in PathMatcher; higher
up the stack, we were then resorting to using a list of PathMatchers.
globset crate exposes a GlobSet type that's better suited for this use
case. In my benchmarks, using a single PathMatcher with GlobSet instead
of a Vec of PathMatchers with Globs is about 3 times faster with the
default 'file_scan_exclusions' values. This slightly improves our
project load time for projects with large # of files, as showcased in
the following videos of loading a project with 100k source files. This
project is *not* a git repository, so it should measure raw overhead on
our side.
Current nightly: 51404d4ea0https://github.com/zed-industries/zed/assets/24362066/e0aa9f8c-aae6-4348-8d42-d20bd41fcd76
versus this PR:
https://github.com/zed-industries/zed/assets/24362066/408dcab1-cee2-4c9e-a541-a31d14772dd7
Release Notes:
- Improved performance in large worktrees
Release Notes:
- Added a more detailed message in place of the generic `checking...`
messages when Rust-analyzer is running.
- Added a rate limit for language server status messages, to reduce
noisiness of those updates.
- Added a `cancel language server work` action which will cancel
long-running language server tasks.
---------
Co-authored-by: Richard <richard@zed.dev>
This PR extracts the definition of the various Zed paths out of `util`
and into a new `paths` crate.
`util` is for generic utils, while these paths are Zed-specific. For
instance, `gpui` depends on `util`, and it shouldn't have knowledge of
these paths, since they are only used by Zed.
Release Notes:
- N/A
I previously put this logic directly into `project.rs`, but it doesn't
feel good to pollute that code with telemetry logic, so I've moved it
over to `telemetry.rs`.
Release Notes:
- N/A
This fixes `project_tests::rescan_and_remote_updates` .
That test was actually correctly failing, revealing two bugs on Linux.
Release Notes:
- Fixed an issue where file renames were not detected on Linux.
- Fixed performance problems caused by excessive file system events on
Linux.
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
This PR disables the
`project::project_tests::test_rescan_and_remote_updates` test on Linux,
as we've been seeing it fail quite consistently in CI.
We can re-enable it once we've had a chance to investigate and fix.
Release Notes:
- N/A
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>
TODO:
- [x] Finish GPUI changes on other operating systems
This is a largely internal change to how we report data to our
diagnostics and telemetry. This PR also includes an update to our blade
backend which allows us to report errors in a more useful way when
failing to initialize blade.
Release Notes:
- N/A
---------
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
This PR adds support for [linked editing of
ranges](https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/#textDocument_linkedEditingRange),
which in short means that editing one part of a file can now change
related parts in that same file. Think of automatically renaming
HTML/TSX closing tags when the opening one is changed.
TODO:
- [x] proto changes
- [x] Allow disabling linked editing ranges on a per language basis.
Fixes#4535
Release Notes:
- Added support for linked editing ranges LSP request. Editing opening
tags in HTML/TSX files (with vtsls) performs the same edit on the
closing tag as well (and vice versa). It can be turned off on a language-by-language basis with the following setting:
```
"languages": {
"HTML": {
"linked_edits": true
},
}
```
---------
Co-authored-by: Bennet <bennet@zed.dev>
This PR is an internal refactor in preparation for remote editing. It
restructures the public interface of `Worktree`, reducing the number of
call sites that assume that a worktree is local or remote.
* The Project no longer calls `worktree.as_local_mut().unwrap()` in code
paths related to basic file operations
* Fewer code paths in the app rely on the worktree's `LocalSnapshot`
* Worktree-related RPC message handling is more fully encapsulated by
the `Worktree` type.
to do:
* [x] file manipulation operations
* [x] sending worktree updates when sharing
for later
* opening buffers
* updating open buffers upon worktree changes
Release Notes:
- N/A
Release Notes:
- Added support for looking up the `rust-analyzer` binary in `$PATH`. This allows using such tools as `asdf` and nix to configure per-folder rust installations. To enable this behavior, use the `path_lookup` key when configuring the `rust-analyzer` `binary`: `{"lsp": {"rust-analyzer": {"binary": {"path_lookup": true }}}}`.
The `worktree` crate mainly provides an in-memory model of a directory
and its git repositories. But because it was originally extracted from
the Project crate, it also contained lingering bits of code that were
outside of that area:
* it had a little bit of logic related to buffers (though most buffer
management lives in `project`)
* it had a *little* bit of logic for storing diagnostics (though the
vast majority of LSP and diagnostic logic lives in `project`)
* it had a little bit of logic for sending RPC message (though the
*receiving* logic for those RPC messages lived in `project`)
In this PR, I've moved those concerns entirely to the project crate
(where they were already dealt with for the most part), so that the
worktree crate can be more focused on its main job, and have fewer
dependencies.
Worktree no longer depends on `client` or `lsp`. It still depends on
`language`, but only because of `impl language::File for
worktree::File`.
Release Notes:
- N/A
Fixes https://github.com/zed-industries/zed/issues/10890
* removes `unwrap()` that caused panics for text elements with no text,
remaining after edit state is cleared but project entries are not
updated, having the fake, "new entry"
* improves discoverability of the FS errors during file/directory
creation: now those are shown as workspace notifications
* stops printing anyhow backtraces in workspace notifications, printing
the more readable chain of contexts instead
* better indicates when new entries are created as excluded ones
Release Notes:
- Improve excluded entry creation workflow in the project panel
([10890](https://github.com/zed-industries/zed/issues/10890))
Previously, each git `Repository` object was held inside of a mutex.
This was needed because libgit2's Repository object is (as one would
expect) not thread safe. But now, the two longest-running git operations
that Zed performs, (`status` and `blame`) do not use libgit2 - they
invoke the `git` executable. For these operations, it's not necessary to
hold a lock on the repository.
In this PR, I've moved our mutex usage so that it only wraps the libgit2
calls, not our `git` subprocess spawns. The main user-facing impact of
this is that the UI is much more responsive when initially opening a
project with a very large git repository (e.g. `chromium`, `webkit`,
`linux`).
Release Notes:
- Improved Zed's responsiveness when initially opening a project
containing a very large git repository.
- Confirming a completion now runs the command immediately
- Hitting `enter` on a line with a command now runs it
- The output of commands gets folded away and replaced with a custom
placeholder
- Eliminated ambient context
<img width="1588" alt="image"
src="https://github.com/zed-industries/zed/assets/482957/b1927a45-52d6-4634-acc9-2ee539c1d89a">
Release Notes:
- N/A
---------
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Fixes https://github.com/zed-industries/zed/issues/9575
Fixes https://github.com/zed-industries/zed/issues/4294
### Problem
When a large git repository's `.git` folder changes (due to a `git
commit`, `git reset` etc), Zed needs to recompute the git status for
every file in that git repository. Part of computing the git status is
the *unstaged* part - the comparison between the content of the file and
the version in the git index. In a large git repository like `chromium`
or `linux`, this is inherently pretty slow.
Previously, we performed this git status all at once, and held a lock on
our `BackgroundScanner`'s state for the entire time. On my laptop, in
the `linux` repo, this would often take around 13 seconds.
When opening a file, Zed always refreshes the metadata for that file in
its in-memory snapshot of worktree. This is normally very fast, but if
another task is holding a lock on the `BackgroundScanner`, it blocks.
### Solution
I've restructured how Zed handles Git statuses, so that when a git
repository is updated, we recompute files' git statuses in fixed-sized
batches. In between these batches, the `BackgroundScanner` is free to
perform other work, so that file operations coming from the main thread
will still be responsive.
Release Notes:
- Fixed a bug that caused long delays in opening files right after
performing a commit in very large git repositories.
This PR adds initial support for defining slash commands for the
Assistant from extensions.
Slash commands are defined in an extension's `extension.toml`:
```toml
[slash_commands.gleam-project]
description = "Returns information about the current Gleam project."
requires_argument = false
```
and then executed via the `run_slash_command` method on the `Extension`
trait:
```rs
impl Extension for GleamExtension {
// ...
fn run_slash_command(
&self,
command: SlashCommand,
_argument: Option<String>,
worktree: &zed::Worktree,
) -> Result<Option<String>, String> {
match command.name.as_str() {
"gleam-project" => Ok(Some("Yayyy".to_string())),
command => Err(format!("unknown slash command: \"{command}\"")),
}
}
}
```
Release Notes:
- N/A
Release Notes:
- Added ZED_RELATIVE_FILE (path to current file relative to worktree
root) and ZED_DIRNAME (path to the directory containing current file)
task variables.
This PR changes the interface of ContextProvider, allowing it to inspect
*all* variables set so far during the process of building
`TaskVariables`. This makes it possible to capture e.g. an identifier in
tree-sitter query, process it and then export it as a task variable.
Notably, the list of variables includes captures prefixed with leading
underscore; they are removed after all calls to `build_context`, but it
makes it possible to capture something and then conditionally preserve
it (and perhaps modify it).
Release Notes:
- N/A
This fixes#4529 by allowing unsaved buffers to be formatted with
prettier.
Steps to do that:
1. Create a new buffer
2. Set language for the buffer (e.g.: `language selector: toggle` and
JSON)
3. In settings, set prettier parser for language (can't be inferred,
since we don't have filename) and allow formatting with prettier:
```json
{
"languages": {
"JSON": {
"prettier": {
"allowed": true,
"parser": "json"
}
}
}
}
```
4. Use `editor: format`
Release Notes:
- Added ability to format unsaved buffers with Prettier. Requirement is
to set a Prettier parser in the user settings. Example for JSON: `{
"languages": { "JSON": { "prettier": { "allowed": true, "parser": "json"
} } } }` ([#4529](https://github.com/zed-industries/zed/issues/4529)).
Demo:
https://github.com/zed-industries/zed/assets/1185253/d24e490b-2e2c-4a5d-95a8-fc8675523780
This fixes#10224 by handling `client/unregisterCapability` requests
that have a `workspace/didChangeWatchedFiles` method.
While debugging the issue, I found out that `gopls` seems to block
indefinitely when there's no reply to the `client/unregisterCapability`
request. Even an empty response would fix the issue.
Seems like gopls 15.x and later seem to handle nested subfolders well,
but do not handle unanswered requests.
Instead of replying with an empty response, I decided to change how we
handle file watching and keep a list of all registered paths so that we
can then unregister paths and recreate the glob patterns.
Release Notes:
- Fixed `gopls` not working correctly when the `go.mod` file was in a
subfolder and not the root folder of the project opened in Zed.
([#10224](https://github.com/zed-industries/zed/issues/10224)).
Fixes#5267
TODO:
- [x] Publish our fork of vscode-langservers-extracted on GH and wire
that through as a language server of choice for HTML extension.
- [x] Figure out how to prevent edits made by remote participants from
moving the cursor of a host.
Release Notes:
- Added support for autoclosing of HTML tags in local projects.
This fixes#10532 by properly making use of `itemDefaults.data` when
that is sent along next to completion `items`.
With this line here we tell the language server that we support `data`
in `itemDefaults`, but we actually never checked for it and never used
it:
a0d7ec9f8e/crates/lsp/src/lsp.rs (L653)
In the case of `tailwindcss-language-server` that means that most of the
items it returns (more than 10k items!) were missing the `data`
attribute, since the language server thought it can send it along in the
`itemDefaults` (because we advertised our capability to use it.)
When we then did a `completionItem/resolve`, we would not send a `data`
attribute along, which lead to an error on the
`tailwindcss-language-server` side and thus no documentation.
This PR also adds support for the other `itemDefaults` that could be
sent along and that we say we support:
a0d7ec9f8e/crates/lsp/src/lsp.rs (L650-L653)
`editRange` we handle separately, so this PR only adds the other 3.
Release Notes:
- Fixed documentation not showing up for completion items coming from
`tailwindcss-language-server`.
([#10532](https://github.com/zed-industries/zed/issues/10532)).
Demo:
https://github.com/zed-industries/zed/assets/1185253/bc5ea0b3-7d83-499f-a908-b0d2a1db8a41
Closes https://github.com/zed-industries/zed/issues/11517
* Removes forced prettier parser name for languages, making `auto`
command to run prettier on every file by default.
* Moves prettier configs away from plugin language declarations into
language settings
Release Notes:
- N/A
Release Notes:
- Improved tasks modal by highlighting a distinction between a task
template and concrete task instance and surfacing available keybindings
more prominently. Task templates are now always available in the modal,
even if there's already a history entry with the same label.
- Changed default key binding for "picker::UseSelectedQuery" to `opt-e`.
I ran into this when trying to get #11550 working: the VCS menu would
open repositories on its owned, based on paths, instead of going through
the worktree on which we already store the git repositories.
Release Notes:
- N/A
Adds a supermaven provider for completions. There are various other
refactors amidst this branch, primarily to make copilot no longer a
dependency of project as well as show LSP Logs for global LSPs like
copilot properly.
This feature is not enabled by default. We're going to seek to refine it
in the coming weeks.
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Co-authored-by: Max <max@zed.dev>
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
As an attempt to do things better when showing diff hunks, store diff
base as Rope, not String, to have cheaper clones when the diff base text
is reused, e.g. creating another buffer with the diff base text for hunk
diff expanding.
Release Notes:
- N/A
Co-Authored-By: Mikayla <mikayla@zed.dev>
In a fit of ill-advisedness I called these things remote projects;
forgetting that remote project is also what we call collaboratively
shared projects.
Release Notes:
- N/A
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
Co-authored-by: Bennet <bennetbo@gmx.de>
This allows function call (i.e. snippet) completion with
`typescript-language-server`. So far that didn't work, because
`typescript-language-server` doesn't respond with `insertText` when
getting the completions, but only when then sending
`completionItem/resolve` requests. See:
https://github.com/hrsh7th/nvim-cmp/issues/646#issuecomment-992765479
What this PR does is to support text edits in the response to
`completionItem/resolve`, which means updating the completion item.
It then enables this feature by default for
`typescript-language-server`.
TODOs:
- [x] Make this work over collab
- [x] Test that this doesn't break existing language server support
- [x] Refactor duplicated code
Release Notes:
- Added support for function call completion when using
`typescript-language-server`. This will result in parameters being
added, which can then be changed and navigated with `<tab>`. For this to
work with `typescript-language-server`, the documentation for a given
completion item needs to be resolved, meaning that if one types very
quickly and accepts completion before `typescript-language-server` could
respond with the documentation, no full function completion is used.
Demo:
https://github.com/zed-industries/zed/assets/1185253/c23ebe12-5902-4b50-888c-d9b8cd32965d
Part of https://github.com/zed-industries/zed/issues/4523
Added two new actions with the default keybindings
```
"cmd-'": "editor::ToggleHunkDiff",
"cmd-\"": "editor::ExpandAllHunkDiffs",
```
that allow to browse git hunk diffs in Zed:
https://github.com/zed-industries/zed/assets/2690773/9a8a7d10-ed06-4960-b4ee-fe28fc5c4768
The hunks are dynamic and alter on user folds and modifications, or
toggle hidden, if the modifications were not adjacent to the expanded
hunk.
Release Notes:
- Added `editor::ToggleHunkDiff` (`cmd-'`) and
`editor::ExpandAllHunkDiffs` (`cmd-"`) actions to browse git hunk diffs
in Zed
I think the previous code was missing a `return` in there because it
always overwrote the `completion.documentation` field, even if the
`text.is_empty()` is true.
Release Notes:
- N/A
TODO:
- [x] Don't immediately seg fault
- [x] Implement for directories
- [x] Add cmd-delete to remove files
- [ ] ~~Add setting for trash vs. delete~~ You can just use keybindings
to change the behavior.
fixes https://github.com/zed-industries/zed/issues/7228
fixes https://github.com/zed-industries/zed/issues/5094
Release Notes:
- Added a new `project_panel::Trash` action and changed the default
behavior for `backspace` and `delete` in the project panel to send a
file to the systems trash, instead of permanently deleting it
([#7228](https://github.com/zed-industries/zed/issues/7228),
[#5094](https://github.com/zed-industries/zed/issues/5094)). The
original behavior can be restored by adding the following section to
your keybindings:
```json5
[
// ...Other keybindings...
{
"context": "ProjectPanel",
"bindings": {
"backspace": "project_panel::Delete",
"delete": "project_panel::Delete",
}
}
]
Still TODO:
* Disable the new save-as for local projects
* Wire up sending the new path to the remote server
Release Notes:
- Added the ability to "Save-as" in remote projects
---------
Co-authored-by: Nathan <nathan@zed.dev>
Co-authored-by: Bennet <bennetbo@gmx.de>
For a long time, we've had problems where diagnostics can end up showing
up inconsistently in different views. This PR is my attempt to prevent
that, and to simplify the system in the process. There are some UX
changes.
Diagnostic behaviors that have *not* changed:
* In-buffer diagnostics update immediately when LSPs send diagnostics
updates.
* The diagnostic counts in the status bar indicator also update
immediately.
Diagnostic behaviors that this PR changes:
* [x] The tab title for the project diagnostics view now simply shows
the same counts as the status bar indicator - the project's current
totals. Previously, this tab title showed something slightly different -
the numbers of diagnostics *currently shown* in the diagnostics view's
excerpts. But it was pretty confusing that you could sometimes see two
different diagnostic counts.
* [x] The project diagnostics view **never** updates its excerpts while
the user might be in the middle of typing it that view, unless the user
expressed an intent for the excerpts to update (by e.g. saving the
buffer). This was the behavior we originally implemented, but has
changed a few times since then, in attempts to fix other issues. I've
restored that invariant.
Times when the excerpts will update:
* diagnostics are updated while the diagnostics view is not focused
* the user changes focus away from the diagnostics view
* the language server sends a `work done progress end` message for its
disk-based diagnostics token (i.e. cargo check finishes)
* the user saves a buffer associated with a language server, and then a
debounce timer expires
* [x] The project diagnostics view indicates when its diagnostics are
stale. States:
* when diagnostics have been updated while the diagnostics view was
focused:
* the indicator shows a 'refresh' icon
* clicking the indicator updates the excerpts
* when diagnostics have been updated, but a file has been saved, so that
the diagnostics will soon update, the indicator is disabled
With these UX changes, the only 'complex' part of the our diagnostics
presentation is the Project Diagnostics view's excerpt management,
because it needs to implement the deferred updates in order to avoid
disrupting the user while they may be typing. I want to take some steps
to reduce the potential for bugs in this view.
* [x] Reduce the amount of state that the view uses, and simplify its
implementation
* [x] Add a randomized test that checks the invariant that a mutated
diagnostics view matches a freshly computed diagnostics view
## Release Notes
- Reworked the project diagnostics view:
- Fixed an issue where the project diagnostics view could update its
excerpts while you were typing in it.
- Fixed bugs where the project diagnostics view could show the wrong
excerpts.
- Changed the diagnostics view to always update its excerpts eagerly
when not focused.
- Added an indicator to the project diagnostics view's toolbar, showing
when diagnostics have been changed.
---------
Co-authored-by: Richard Feldman <oss@rtfeldman.com>
This PR adds the ability for the Tailwind language server
(`tailwindcss-language-server`) to be controlled by the
`language_servers` setting.
Now in your settings you can indicate that the Tailwind language server
should be used for a given language, even if that language does not have
the Tailwind language server registered for it already:
```json
{
"languages": {
"My Language": {
"language_servers": ["tailwindcss-language-server", "..."]
}
}
}
```
Release Notes:
- N/A
This PR fixes the way we select the primary language server for use with
formatting.
Previously we were just taking the first one in the list, but this could
be the wrong one in cases where a language server was provided by an
extension in conjunction with a built-in language server (e.g.,
Tailwind).
We now use the `primary_language_server_for_buffer` method to more
accurately identify the primary one.
Fixes https://github.com/zed-industries/zed/issues/10902.
Release Notes:
- Fixed an issue where the wrong language server could be used for
formatting.
This PR adds a new log message indicating which language servers will be
started for a given language.
The aim is to make debugging the usage of the new `language_servers`
setting (#10911) easier.
Release Notes:
- N/A
This PR adds a new `language_servers` setting underneath the language
settings.
This setting controls which of the available language servers for a
given language will run.
The `language_servers` setting is an array of strings. Each item in the
array must be either:
- A language server ID (e.g., `"rust-analyzer"`,
`"typescript-language-server"`, `"eslint"`, etc.) denoting a language
server that should be enabled.
- A language server ID prefixed with a `!` (e.g., `"!rust-analyzer"`,
`"!typescript-language-server"`, `"!eslint"`, etc.) denoting a language
server that should be disabled.
- A `"..."` placeholder, which will be replaced by the remaining
available language servers that haven't already been mentioned in the
array.
For example, to enable the Biome language server in place of the default
TypeScript language server, you would add the following to your
settings:
```json
{
"languages": {
"TypeScript": {
"language_servers": ["biome", "!typescript-language-server", "..."]
}
}
}
```
More details can be found in #10906.
Release Notes:
- Added `language_servers` setting to language settings for customizing
which language server(s) run for a given language.
Release Notes:
- Made remote projects per-user instead of per-channel. If you'd like to
be part of the remote development alpha, please email hi@zed.dev.
---------
Co-authored-by: Bennet Bo Fenner <53836821+bennetbo@users.noreply.github.com>
Co-authored-by: Bennet <bennetbo@gmx.de>
Co-authored-by: Nate Butler <1714999+iamnbutler@users.noreply.github.com>
Co-authored-by: Nate Butler <iamnbutler@gmail.com>
Part of https://github.com/zed-industries/zed/issues/5141
* adds "run selection" and "run file" tasks for bash and Python.
* replaces newlines with `\n` symbols in the human-readable task labels
* properly escapes task command arguments when spawning the task in
terminal
Caveats:
* bash tasks will always use user's default shell to spawn the
selections, but they should rather respect the shebang line even if it's
not selected
* Python tasks will always use `python3` to spawn its tasks now, as
there's no proper mechanism in Zed to deal with different Python
executables
Release Notes:
- Added tasks for bash and Python to execute selections and open files
in terminal
This adds so-called "inline git blame" to the editor that, when turned
on, shows `git blame` information about the current line inline:
When the inline information is hovered, a new tooltip appears that
contains more information on the current commit:
The commit message in this tooltip is rendered as Markdown, is
scrollable and clickable.
The tooltip is now also the tooltip used in the gutter:
## Settings
1. The inline git blame information can be turned on and off via
settings:
```json
{
"git": {
"inline_blame": {
"enabled": true
}
}
}
```
2. Optionally, a delay can be configured. When a delay is set, the
inline blame information will only show up `x milliseconds` after a
cursor movement:
```json
{
"git": {
"inline_blame": {
"enabled": true,
"delay_ms": 600
}
}
}
```
3. It can also be turned on/off for the current buffer with `editor:
toggle git blame inline`.
## To be done in follow-up PRs
- [ ] Add link to pull request in tooltip
- [ ] Add avatars of users if possible
## Release notes
Release Notes:
- Added inline `git blame` information the editor. It can be turned on
in the settings with `{"git": { "inline_blame": "on" } }` for every
buffer or, temporarily for the current buffer, with `editor: toggle git
blame inline`.
This introduces semantic indexing in Zed based on chunking text from
files in the developer's workspace and creating vector embeddings using
an embedding model. As part of this, we've created an embeddings
provider trait that allows us to work with OpenAI, a local Ollama model,
or a Zed hosted embedding.
The semantic index is built by breaking down text for known
(programming) languages into manageable chunks that are smaller than the
max token size. Each chunk is then fed to a language model to create a
high dimensional vector which is then normalized to a unit vector to
allow fast comparison with other vectors with a simple dot product.
Alongside the vector, we store the path of the file and the range within
the document where the vector was sourced from.
Zed will soon grok contextual similarity across different text snippets,
allowing for natural language search beyond keyword matching. This is
being put together both for human-based search as well as providing
results to Large Language Models to allow them to refine how they help
developers.
Remaining todo:
* [x] Change `provider` to `model` within the zed hosted embeddings
database (as its currently a combo of the provider and the model in one
name)
Release Notes:
- N/A
---------
Co-authored-by: Nathan Sobo <nathan@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Conrad Irwin <conrad@zed.dev>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
Co-authored-by: Antonio <antonio@zed.dev>
This reverts commit caed275fbf.
NOTE: this should not be merged until #9668 is on stable and the
`ZedVersion#can_collaborate` is updated to exclude all clients without
that change.
Release Notes:
- N/A
---------
Co-authored-by: Piotr Osiewicz <24362066+osiewicz@users.noreply.github.com>
<img width="1035" alt="Screenshot 2024-04-11 at 13 13 44"
src="https://github.com/zed-industries/zed/assets/13402668/cd0e96a0-41c6-4757-8840-97d15a75c511">
Release Notes:
- Added a notification to show possible `git blame` errors if it fails to run.
Caveats:
- ~git blame now executes in foreground
executor (required since the Fut is !Send)~
TODOs:
- After a failed toggle, the app thinks the blame
is shown. This means toggling again will do nothing
instead of retrying. (Caused by editor.show_git_blame
being set to true before the git blame is generated)
- ~(Maybe) Trim error?~ Done
---------
Co-authored-by: Thorsten Ball <mrnugget@gmail.com>
New list (used tasks are above the separator line, sorted by the usage
recency), then all language tasks, then project-local and global tasks
are listed.
Note that there are two test tasks (for `test_name_1` and `test_name_2`
functions) that are created from the same task template:
<img width="563" alt="Screenshot 2024-04-10 at 01 00 46"
src="https://github.com/zed-industries/zed/assets/2690773/7455a82f-2af2-47bf-99bd-d9c5a36e64ab">
Tasks are deduplicated by labels, with the used tasks left in case of
the conflict with the new tasks from the template:
<img width="555" alt="Screenshot 2024-04-10 at 01 01 06"
src="https://github.com/zed-industries/zed/assets/2690773/8f5a249e-abec-46ef-a991-08c6d0348648">
Regular recent tasks can be now removed too:
<img width="565" alt="Screenshot 2024-04-10 at 01 00 55"
src="https://github.com/zed-industries/zed/assets/2690773/0976b8fe-b5d7-4d2a-953d-1d8b1f216192">
When the caret is in the place where no function symbol could be
retrieved, no cargo tests for function are listed in tasks:
<img width="556" alt="image"
src="https://github.com/zed-industries/zed/assets/2690773/df30feba-fe27-4645-8be9-02afc70f02da">
Part of https://github.com/zed-industries/zed/issues/10132
Reworks the task code to simplify it and enable proper task labels.
* removes `trait Task`, renames `Definition` into `TaskTemplate` and use
that instead of `Arc<dyn Task>` everywhere
* implement more generic `TaskId` generation that depends on the
`TaskContext` and `TaskTemplate`
* remove `TaskId` out of the template and only create it after
"resolving" the template into the `ResolvedTask`: this way, task
templates, task state (`TaskContext`) and task "result" (resolved state)
are clearly separated and are not mixed
* implement the logic for filtering out non-related language tasks and
tasks that have non-resolved Zed task variables
* rework Zed template-vs-resolved-task display in modal: now all reruns
and recently used tasks are resolved tasks with "fixed" context (unless
configured otherwise in the task json) that are always shown, and Zed
can add on top tasks with different context that are derived from the
same template as the used, resolved tasks
* sort the tasks list better, showing more specific and least recently
used tasks higher
* shows a separator between used and unused tasks, allow removing the
used tasks same as the oneshot ones
* remote the Oneshot task source as redundant: all oneshot tasks are now
stored in the inventory's history
* when reusing the tasks as query in the modal, paste the expanded task
label now, show trimmed resolved label in the modal
* adjusts Rust and Elixir task labels to be more descriptive and closer
to bash scripts
Release Notes:
- Improved task modal ordering, run and deletion capabilities
This PR renames `language::Buffer::new` to `language::Buffer::local` and
simplifies its interface. Instead of taking a replica id (which should
always be 0 for the local case) and a `BufferId`, which was awkward and
verbose to construct, it simply takes text and a `cx`.
It uses the `cx` to derive a `BufferId` from the `EntityId` associated
with the `cx`, which should always be positive based on the following
analysis...
We convert the entity id to a u64 using this method on `EntityId`, which
is defined by macros in the `slotmap` crate:
```rust
pub fn as_ffi(self) -> u64 {
(u64::from(self.version.get()) << 32) | u64::from(self.idx)
}
```
If you look at the type of `version` in `KeyData`, it is non-zero:
```rust
#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub struct KeyData {
idx: u32,
version: NonZeroU32,
}
```
This commit also adds `Context::reserve_model` and
`Context::insert_model` to determine a model's entity ID before it is
created, which we need in order to assign a `BufferId` in the background
when loading a buffer asynchronously.
Release Notes:
- N/A
---------
Co-authored-by: Piotr Osiewicz <24362066+osiewicz@users.noreply.github.com>
