This PR introduces a separate backend service for making LLM calls.
It exposes an HTTP interface that can be called by Zed clients. To call
these endpoints, the client must provide a `Bearer` token. These tokens
are issued/refreshed by the collab service over RPC.
We're adding this in a backwards-compatible way. Right now the access
tokens can only be minted for Zed staff, and calling this separate LLM
service is behind the `llm-service` feature flag (which is not
automatically enabled for Zed staff).
Release Notes:
- N/A
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This UI change is behind a `ZedPro` feature flag so that it won't be
visible until we're ready to launch that service.
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
Co-authored-by: Marshall <marshall@zed.dev>
This PR adds support for a new encryption format for exchanging access
tokens during the authentication flow.
The new format uses Optimal Asymmetric Encryption Padding (OAEP) instead
of PKCS#1 v1.5, which is known to be vulnerable to side-channel attacks.
**Note: We are not yet encrypting access tokens using the new format, as
this is a breaking change between the client and the server. This PR
only adds support for it, and makes it so the client and server can
decrypt either format moving forward.**
This required bumping the RSA key size from 1024 bits to 2048 bits. This
is necessary to be able to encode the access token into the ciphertext
when using OAEP.
This also follows OWASP recommendations:
> If ECC is not available and RSA must be used, then ensure that the key
is at least 2048 bits.
>
> —
[source](https://cheatsheetseries.owasp.org/cheatsheets/Cryptographic_Storage_Cheat_Sheet.html#algorithms)
Release Notes:
- N/A
This PR upgrades the `rsa` crate to v0.9.6.
The version we were using was rather old, and for something
security-sensitive we should be using a recent version.
No behavioral changes have been made, just updates to account for
changes in the crate's API.
Release Notes:
- N/A
This PR adds some tests to ensure we don't regress in our public key
encoding/decoding capabilities when making changes in this area.
Release Notes:
- N/A
This is a ~small~ pure refactor that's a step toward SSH remoting. I've
extracted the Project's buffer state management into a smaller, separate
struct called `BufferStore`, currently in the same crate. I did this as
a separate PR to reduce conflicts between main and `remoting-over-ssh`.
The idea is to make use of this struct (and other smaller structs that
make up `Project`) in a dedicated, simpler `HeadlessProject` type that
we will use in the SSH server to model the remote end of a project. With
this approach, as we develop the headless project, we can avoid adding
more conditional logic to `Project` itself (which is already very
complex), and actually make `Project` a bit smaller by extracting out
helper objects.
Release Notes:
- N/A
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
#### Lazily loading channels
I've added a new RPC message called `SubscribeToChannels` that the
client now sends when it first renders the channels panel. This causes
the server to load the channels for that client and send updates to that
client as channels are updated. Previously, the server did this upon
connection.
For backwards compatibility, the server will inspect clients' version,
and continue to do this work immediately for old clients.
#### Optimizations
Running collab locally, I realized that upon connecting, we were running
two concurrent transactions that *both* queried the `channel_members`
table: one for loading your channels, and one for loading your channel
invites. I've combined these into one query. In addition, we now use a
join to load channels + members, as opposed to two separate queries.
Even though `where id in` is efficient, it adds an extra round trip to
the database, keeping the transaction open for slightly longer.
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>
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 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>
@ConradIrwin This is a followup for #9035 as agreed.
Release Notes:
- Fixed mention notifications are updated when channel message is
updated. And mention notifications are removed when message is removed.
---------
Co-authored-by: Bennet Bo Fenner <53836821+bennetbo@users.noreply.github.com>
This adds a new action to the editor: `editor: toggle git blame`. When
used it turns on a sidebar containing `git blame` information for the
currently open buffer.
The git blame information is updated when the buffer changes. It handles
additions, deletions, modifications, changes to the underlying git data
(new commits, changed commits, ...), file saves. It also handles folding
and wrapping lines correctly.
When the user hovers over a commit, a tooltip displays information for
the commit that introduced the line. If the repository has a remote with
the name `origin` configured, then clicking on a blame entry opens the
permalink to the commit on the code host.
Users can right-click on a blame entry to get a context menu which
allows them to copy the SHA of the commit.
The feature also works on shared projects, e.g. when collaborating a
peer can request `git blame` data.
As of this PR, Zed now comes bundled with a `git` binary so that users
don't have to have `git` installed locally to use this feature.
### Screenshots
### TODOs
- [x] Bundling `git` binary
### Release Notes
Release Notes:
- Added `editor: toggle git blame` command that toggles a sidebar with
git blame information for the current buffer.
---------
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Piotr <piotr@zed.dev>
Co-authored-by: Bennet <bennetbo@gmx.de>
Co-authored-by: Mikayla <mikayla@zed.dev>
* Store extensions versions' wasm API version in the database
* Share a common struct for extension API responses between collab and
client
* Add wasm API version and schema version to extension API responses
Release Notes:
- N/A
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
Resurrected this from some assistant work I did in Spring of 2023.
- [x] Resurrect streaming responses
- [x] Use streaming responses to enable AI via Zed's servers by default
(but preserve API key option for now)
- [x] Simplify protobuf
- [x] Proxy to OpenAI on zed.dev
- [x] Proxy to Gemini on zed.dev
- [x] Improve UX for switching between openAI and google models
- We current disallow cycling when setting a custom model, but we need a
better solution to keep OpenAI models available while testing the google
ones
- [x] Show remaining tokens correctly for Google models
- [x] Remove semantic index
- [x] Delete `ai` crate
- [x] Cloud front so we can ban abuse
- [x] Rate-limiting
- [x] Fix panic when using inline assistant
- [x] Double check the upgraded `AssistantSettings` are
backwards-compatible
- [x] Add hosted LLM interaction behind a `language-models` feature
flag.
Release Notes:
- We are temporarily removing the semantic index in order to redesign it
from scratch.
---------
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Thorsten <thorsten@zed.dev>
Co-authored-by: Max <max@zed.dev>
We couldn't reproduce the panic, but I believe it was possible when
uninstalling an extension while one if its grammars was still loading.
Release Notes:
- Fixed a crash that could happen when uninstalling a language extension
while its grammar was loading.
---------
Co-authored-by: Conrad <conrad@zed.dev>
- Open channel notes and chat on channel click
- WIP
- Fix compile error
- Don't join live kit until requested
- Track in_call state separately from in_room
Release Notes:
- Improved channels: you can now be in a channel without joining the
audio call automatically
**or**
- N/A
---------
Co-authored-by: Nathan Sobo <nathan@zed.dev>
We noticed the following message in my logs when trying to debug some
lag when collaborating:
```
2024-02-06T09:42:09-08:00 [ERROR] error handling message. client_id:3, sender_id:Some(PeerId { owner_id: 327, id: 1123430 }), type:GetCompletions, error:no such connection: 0/0
```
That `0/0` looks like a bogus connection id, constructed via a derived
`Default`. We didn't ever find a code path that would *use* a default
`ConnectionId` and lead to this error, but it did seem like an
improvement to not have a `Default` for that type.
Release Notes:
- N/A
Co-authored-by: Marshall <marshall@zed.dev>
- Disallow sharing gitignored files through collab
- Show errors when failing to open files
- Show a warning to followers when view is unshared
/cc @mikaylamaki, let's update this to use your `private_files` config
before merge.
Release Notes:
- Added the ability to prevent sharing private files over collab.
---------
Co-authored-by: Piotr <piotr@zed.dev>
Co-authored-by: Mikayla <mikayla@zed.dev>
