- Render the output only when needed, fixes the duplicate output that
can happen after opening a saved notebook.
- Vim in Jupyter View with j/k navigation across notebook cells
Release Notes:
- N/A
Closes#15196, #46918
- fix: notebook_ui, use buffer so that notebooks open in remote/WSL
settings.
- fix: add musl in nix for cross-compilation, without this remote server
doesn't build inside NixOS
Release Notes:
- Implement WSL and SSH remote kernels (crates/repl/src/kernels/*) and
wire up spawn/kill kernel proto messages and client requests.
Closes#15947
This adds `repl:ClearCurrentOutput` and `repl:ClearOutputs` commands. No
keybindings are set for this. Just an action people can bind.
Release Notes:
- Added ability to clear outputs by action
Replaces a bunch of `impl FnMut` parameters with `&mut dyn FnMut` for
functions where this is the sole generic parameter.
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Here's some backstory:
* on macOS, @cole-miller and I noticed that since roughly Oct 2025, due
to some changes to latest macOS Tahoe, for any spawned child process we
needed to reset Mach exception ports
(https://github.com/zed-industries/zed/issues/36754 +
https://github.com/RemiKalbe/zed/commit/6e8f2d2ebe5a93753625a3026aeb996de8cb436b)
* the changes in that PR achieve that via `pre_exec` hook on
`std::process::Command` which then abandons `posix_spawn` syscall for
`fork` + `execve` dance on macOS (we tracked it down in Rust's std
implementation)
* as it turns out, `fork` + `execve` is pretty expensive on macOS
(apparently way more so than on other OSes like Linux) and `fork` takes
a process-wide lock on the allocator which is bad
* however, since we wanna reset exception ports on the child, the only
official way supported by Rust's std is to use `pre_exec` hook
* posix_spawn on macOS exposes this tho via a macOS specific extension
to that syscall `posix_spawnattr_setexceptionports_np` but there is no
way to use that via any standard interfaces in `std::process::Command`
* thus, it seemed like a good idea to instead create our own custom
Command wrapper that on non-macOS hosts is a zero-cost wrapper of
`smol::process::Command`, while on macOS we reimplement the minimum to
achieve `smol::process::Command` with `posix_spawn` under-the-hood
Notably, this changeset improves git-blame in very large repos
significantly.
Release Notes:
- Fixed performance spawning child processes on macOS by always forcing
`posix_spawn` no matter what.
---------
Co-authored-by: Cole Miller <cole@zed.dev>
Closes#21529
Show a toast when a message from a Jupyter kernel doesn't parse
properly.
Discovered that some kernels don't include their execution count in an
`ExecuteReply` which is required in the jupyter protocol. Upstream I'll
go make that field either be an `Option` or just stick
`#[serde(default)]` on it so it's not a change in the interface.
Release Notes:
- Show error when parsing a message from a Jupyter kernel fails
Hot on the heels of https://github.com/zed-industries/zed/pull/48817 I'm
bringing the best improvement to the repl underneath: `split()`-able
sockets! Much more will be unlocked by having this.
This split the shell and control `DealerSocket` connections into
independent send/recv halves using the new `split()` API from zeromq
0.5.0 and runtimelib 1.x. This also nicely cleaned things up so we could
have a single `select!` loop over iopub, shell, and control recv halves.
That replaces three separate recv tasks.
This likely closes some issues for certain kernels that would get stuck
either during startup or other flows due to them not sending replies to
specific requests. I'll see if I can find issues around this and update
the release notes after.
This allows us to unlock some nifty new things we can do on the shell
socket, particularly autocompletion for in-memory values, stdin support,
and others. I _think_ it also help with sending and receving
`KernelInfo`, which not all kernels do properly at the start. This makes
us a bit more resilient to errant kernels.
Release Notes:
- N/A
Closes#17985Closes#21911
Force the shutdown of the kernel by ensuring the kernel sessions are
dropped on app quit.
Release Notes:
- Fixed shutdown of kernels on app exit
Bump the runtimed ecosystem packages to their 1.0 releases:
- `jupyter-protocol`: 0.10.0 → 1.1.0
- `jupyter-websocket-client`: 0.15.0 → 1.0.0
- `nbformat`: 0.15.0 → 1.0.0
- `runtimelib`: 0.30.0 → 1.1.0
One breaking change: `MimeType::Json` now wraps `serde_json::Value`
directly instead of `serde_json::Map<String, Value>`, so the redundant
`Value::Object(...)` wrapping in `outputs.rs` was removed.
Closes#41649
More Quality of Life improvements and Bug Fixes are coming in new PRs
after this is merged.
Release Notes:
- Fixes startup for R kernels and Python kernels on windows.
The new multi workspace introduced in #47795 changed the window root
from `Workspace` to `MultiWorkspace`, which broke
`Workspace::for_window()` (assuming that was meant to). That returns
`None` now. The REPL action registration in `repl_sessions_ui.rs` used
this to check if the project was local, so when it got None, it silently
skipped registering `repl::Run` and `repl::RunInPlace` on every editor.
Luckily we can just get the project directly from the editor in order to
register actions.
Release Notes:
- N/A
- Keyboard navigation where you can traverse through cells using up and
down arrow
- Jupyter Logo added
- Initialize kernel as shutdown for more predictable behavior
- Ability to create .ipynb files with bare essential metadata.
- Optimize editor initialization to avoid cloning the entire notebook
and shortcuts
Release Notes:
- N/A
---------
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
- This also sends the cursor to block placement anchor which is the
standard thing to happen when we run cmd/ctrl + shift + enter, this is
usually used for Run and Move onto next cell.
- Perhaps the ability to stay on the same code will be tackled on
further works where not using the shift modifier would signify stay and
"just" run the cell. Like #46868Closes#48069
Release Notes:
- Fixed cursor becoming invisible on the last line of REPL cells after
re-running
Closes#47113
Adds configurable REPL output size limits with two new settings,
`repl.output_max_height_lines` and `repl.output_max_width_columns`, so
large outputs scroll instead of expanding and images scale down to fit
the available space. The output containers in both inline REPL blocks
and notebook cells now respect these bounds, and image sizing uses the
same text metrics as the terminal output for consistent column-based
width calculations.
Release Notes:
- REPL output now supports configurable max height and width limits,
with large outputs scrolling and images scaling to stay within the
viewport.
Brought the Markdown output up to date with how Markdown is used in the
Agent panel. This fixed an issue with outputs that were too large for
the execution view as well as made sure that markdown would wrap.
<img width="3222" height="2334" alt="image"
src="https://github.com/user-attachments/assets/c65efa53-b792-4529-909a-9117053e30be"
/>
Release Notes:
- N/A
Coming from discussion on #25936 and
https://github.com/zed-industries/zed/pull/19756 .
This PR introduces the KernelSession abstraction and adds basic notebook
cell execution inside NotebookEditor.
The following provides a base starter for the development on Notebooks.
What this PR includes today:
Release Notes:
- KernelSession trait extracted. Both REPL and NotebookEditor now share
the same routing mechanism.
- NotebookEditor can launch kernels, execute code cells, and display
outputs.
- Basic cell operations: run current cell, run all, move up/down, add
code/markdown blocks. Keybindings follow Jupyter defaults.
Next Steps:
- [x] Editing support for markdown and code blocks.
- Buffer integration instead of temporary cell text.
- [x] Proper focus behavior when executing or adding cells.
- Kernel control UI.
A little far fetched steps:
- Vim Support
- Cell Handling Improvement and other convenient features and design
from other editors
- Ability to have better parsing for AI Support.
I have attached a video of showcasing some of the features:
https://github.com/user-attachments/assets/37e6f3e5-2022-45f0-a73d-2dd01ebc2932
This PR improves the performance of the `CopyButton` component by
tracking the copied state locally through a `CopyButtonState` struct
instead of making an OS call every time the component re-renders. Also
pushing a slight improvement here by resetting the state after two
seconds so as to make the check mark go away after you clicked to copy.
Release Notes:
- N/A
## Motivation
This PR unifies the async execution infrastructure between GPUI and
other components that depend on the `scheduler` crate (such as our cloud
codebase). By having a scheduler that lives independently of GPUI, we
can enable deterministic testing across the entire stack - testing GPUI
applications alongside cloud services with a single, unified scheduler.
## Summary
This PR completes the integration of the `scheduler` crate into GPUI,
unifying async execution and enabling deterministic testing of GPUI
combined with other components that depend on the scheduler crate.
## Key Changes
### Scheduler Integration (Phases 1-5, previously completed)
- `TestDispatcher` now delegates to `TestScheduler` for timing, clock,
RNG, and task scheduling
- `PlatformScheduler` implements the `Scheduler` trait for production
use
- GPUI executors wrap scheduler executors, selecting `TestScheduler` or
`PlatformScheduler` based on environment
- Unified blocking logic via `Scheduler::block()`
### Dead Code Cleanup
- Deleted orphaned `crates/gpui/src/platform/platform_scheduler.rs`
(older incompatible version)
## Intentional Removals
### `spawn_labeled` and `deprioritize` removed
The `TaskLabel` system (`spawn_labeled`, `deprioritize`) was removed
during this integration. It was only used in a few places for test
ordering control.
cc @maxbrunsfeld @as-cii - The new priority-weighted scheduling in
`TestScheduler` provides similar functionality through
`Priority::High/Medium/Low`. If `deprioritize` is important for specific
test scenarios, we could add it back to the scheduler crate. Let me know
if this is blocking anything.
### `start_waiting` / `finish_waiting` debug methods removed
Replaced by `TracingWaker` in `TestScheduler` - run tests with
`PENDING_TRACES=1` to see backtraces of pending futures when parking is
forbidden.
### Realtime Priority removed
The realtime priority feature was unused in the codebase. I'd prefer to
reintroduce it when we have an actual use case, as the implementation
(bounded channel with capacity 1) could potentially block the main
thread. Having a real use case will help us validate the design.
## Testing
- All GPUI tests pass
- All scheduler tests pass
- Clippy clean
## Architecture
```
┌─────────────────────────────────────────────────────────────┐
│ GPUI │
│ ┌──────────────────────┐ ┌────────────────────────────┐ │
│ │ gpui::Background- │ │ gpui::ForegroundExecutor │ │
│ │ Executor │ │ - wraps scheduler:: │ │
│ │ - scheduler: Arc< │ │ ForegroundExecutor │ │
│ │ dyn Scheduler> │ └────────────┬───────────────┘ │
│ └──────────┬───────────┘ │ │
│ │ │ │
│ └──────────┬──────────────────┘ │
│ ▼ │
│ ┌───────────────────────┐ │
│ │ Arc<dyn Scheduler> │ │
│ └───────────┬───────────┘ │
│ ┌──────────────┴──────────────┐ │
│ ▼ ▼ │
│ ┌──────────────────┐ ┌────────────────────┐ │
│ │ PlatformScheduler│ │ TestScheduler │ │
│ │ (production) │ │ (deterministic) │ │
│ └──────────────────┘ └────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
```
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Yara <git@yara.blue>
Co-authored-by: Zed Zippy <234243425+zed-zippy[bot]@users.noreply.github.com>
Additionally this extracts more functionality into the RunnableVariant which is renamed to GpuiRunnable.
Release Notes:
- N/A
---------
Co-authored-by: Lukas Wirth <lukas@zed.dev>
Co-authored-by: Cole Miller <cole@zed.dev>
Adds various useful things to the repl inspired by ipynb and the julia
vscode extension which can be best seen with this video:

https://github.com/user-attachments/assets/6589715e-3783-456c-8f4b-e2d5a1c4090d
To summarize:
## Inline outputs
Added small, single-line outputs displayed inline at the end of the code
line instead of in a separate block. This provides a cleaner, more
compact view for simple results like numbers or short strings. This
occurs for execution views who only output a single mimetype/plain OR
output nothing, otherwise the default behavior of creating a block will
occur.
It looks like this:
<img width="258" height="35" alt="image"
src="https://github.com/user-attachments/assets/ccdeca3f-c3b7-4387-a4de-53d8b9a25132"
/>
or with a Output
<img width="346" height="55" alt="image"
src="https://github.com/user-attachments/assets/0b4effc9-1bd7-4e8c-802f-8733cdcc77d1"
/>
This was inspired by julia vscode extension, but now it can be used with
any replanguage! Hooray!
<img width="524" height="450" alt="image"
src="https://github.com/user-attachments/assets/a3551e51-f5f7-4d3e-994a-213c9d2f948c"
/>
It saves lots of space compared to the ugly and distracting:
<img width="531" height="546" alt="image"
src="https://github.com/user-attachments/assets/7cf65bae-8ec1-4279-ab19-f0d4ec4052a2"
/>
## Gutters and execution numbers
Added gutters + execution number to display exactly what was executed.
The gutter highlighting is useful for when selecting multiple cells
manually to run, but you dont remember which ones
Ran at different times:
<img width="257" height="58" alt="image"
src="https://github.com/user-attachments/assets/6002ab16-156a-4598-9964-5a6b188e989c"
/>
Ran together:
<img width="306" height="64" alt="image"
src="https://github.com/user-attachments/assets/2690ea35-2bd3-4207-b039-6c0f98dad6e4"
/>
The execution number is useful in the same way that a normal jupyter
notebook execution number is useful.
If a gutter-region does not have a block assigned to it, when you edit
the text in the gutter region, the gutter will disappear, which is
useful for telling when you have modified your code, but does not delete
useful experiment results in blocks:
<img width="280" height="38" alt="image"
src="https://github.com/user-attachments/assets/d7f29224-87e4-4c14-8d9f-41cb10ab5009"
/>
<img width="254" height="31" alt="image"
src="https://github.com/user-attachments/assets/586c9e1d-f53c-4973-affb-c8ca05a7563b"
/>
<img width="264" height="29" alt="image"
src="https://github.com/user-attachments/assets/f306c364-1c92-44bd-9050-ecce1b7822a0"
/>
## Skip empty line
This is a minor fix which is intended to make lab workflow less tedious.
Currently when you execute on an empty line (which might be there for
formatting purposes) nothing will occur. This PR adds the ability to,
when executing from an empty line, skip ahead the range of inclusion
until you reach actual code, and then execute.
Before:
```
code //run execute
//empty space, so you have to move your cursor down or use arrow key
code //run execute
code //run execute
```
After:
```
code //run execute
//empty space, you can now run execute on it and it will include the next line of code
//empty space
code //automatically executed
code //run execute
```
Currently the only piece of tested code is related to this, i still have
to write tests for the gutter annotation api i added and all of the
gutter + inline related code. Also still have to add more config for
this stuff.
@rgbkrk would appreciate a review :D
Closes#22678
Release Notes:
- repl: Added an inline display of execution results (as opposed to the
large execution view) for simple REPL cells
- repl: Improved how execution of empty lines are handled
- repl: Added gutter execution display
There were several places adding a copy icon button, so thought of
encapsulating the logic to copy a given string into the clipboard (and
other small details like swapping the icon and tooltip if copied) into a
component, making it easier to introduce this sort of functionality in
the future, with fewer lines of code.
All it takes (for the simplest case) is:
```rs
CopyButton::new(your_message)
```
<img width="600" height="714" alt="Screenshot 2025-12-29 at 10 50@2x"
src="https://github.com/user-attachments/assets/e6949863-a056-4855-82d8-e4ffb5d62c90"
/>
Release Notes:
- N/A
Improves the scheduler by allowing tasks to have a set priority which
will significantly improve responsiveness.
Release notes:
- N/A
---------
Co-authored-by: Yara <git@yara.blue>
Co-authored-by: dvdsk <noreply@davidsk.dev>
Improves the scheduler by allowing tasks to have a set priority which
will significantly improve responsiveness.
Release notes:
- N/A
---------
Co-authored-by: Yara <git@yara.blue>
Implements a specialized constructor `LanguageName::new_static` for
`&'static str` which reduces allocations.
`LanguageName::new` always backs the underlying `SharedString` with an
owned `Arc<str>` even when a `&'static str` is passed. This makes us
allocate each time we create a new `LanguageName` no matter what.
Creating a specialized constructor for `&'static str` allows us to
essentially construct them for free.
Additional change:
Encourages using explicit constructors to avoid needless allocations.
Currently there were no instances of this trait being called where the
lifetime was not `'static` saving another 48 locations of allocation.
```rust
impl<'a> From<&'a str> for LanguageName {
fn from(str: &'a str) -> Self {
Self(SharedString::new(str))
}
}
// to
impl From<&'static str> for LanguageName {
fn from(str: &'static str) -> Self {
Self(SharedString::new_static(str))
}
}
```
Release Notes:
- N/A
Gist is we only need to block the foreground thread for reparsing if
immediate language changes are useful to the user. That is usually only
the case when they edit the buffer
Release Notes:
- Improved performance of large project searches and project diffs
Co-authored by: David Kleingeld <david@zed.dev>
This PR introduces a new `MultiBufferOffset` new type wrapping size. The
goal of this is to make it clear at the type level when we are
interacting with offsets of a multi buffer versus offsets of a language
/ text buffer. This improves readability of things quite a bit by making
it clear what kind of offsets one is working with while also reducing
accidental bugs by using the wrong kin of offset for the wrong API.
This PR also uncovered two minor bugs due to that.
Does not yet introduce the MultiBufferPoint equivalent, that is for a
follow up PR.
Release Notes:
- N/A *or* Added/Fixed/Improved ...
Closes#40888
This updates runtimed to the latest version, which handles the
"starting" variant of `execution_state`. It actually handles a bunch of
other variants that are not documented in the protocol (see
https://jupyter-client.readthedocs.io/en/stable/messaging.html#kernel-status),
like "starting", "terminating", etc. I added implementations for these
variants as well.
Release Notes:
- Fixed issue that prevented the Ark kernel from working in Zed
(#40888).
---------
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
Release Notes:
- N/A
When I was implementing Input, I often used `TextRun`, but `background`,
`underline` and `strikethrough` were often not used.
So make change to simplify it.
MostlyK