[corgi](https://github.com/ConradIrwin/corgi) is a cargo-compatible build tool that runs `build.rs` scripts and proc-macros in a sandbox (no network, no ambient env, reads limited to a package's own `.rs` sources unless declared). This makes Zed buildable under it. Cargo builds (dev and release) are unaffected. ### Changes - **`scratch` patch**: patch `scratch` to a small local copy (`corgi-patches/scratch`) that reads `OUT_DIR` at runtime instead of baking it in at compile time. `cxx-build` writes its shared cxxbridge headers into `scratch::path(...)`; upstream returns a single global dir (outside any action's `OUT_DIR`) that the sandbox can't grant, while this patch gives each build script a private, writable dir in its own entry. webrtc-sys is self-contained, so per-crate scratch dirs are sufficient. Kept local rather than a cxx git fork because a git checkout of cxx needs symlink support that Windows cargo CI lacks. - **Dev asset loading**: add `util::dev_fs_embed!` and switch dev-mode `rust_embed` asset sources (`assets`, `settings`, `grammars`, `agent`, `edit_prediction_cli`) to read from the checkout at runtime, locating the repo root by walking from the executable up to the enclosing `.git`. This avoids baking `CARGO_MANIFEST_DIR` into the binary (corgi rejects artifacts that embed the build-time checkout path, and it's wrong in any other worktree). Release builds still embed via `#[derive(RustEmbed)]`. - **Stop baking checkout paths into artifacts**: drop `gpui::GPUI_MANIFEST_DIR`; resolve `../gpui` from `gpui_apple`'s build script; stage the metal shader into `OUT_DIR` before compiling so the `.metallib` records the output dir, not the checkout; resolve the repo root at runtime in `remote` and `inspector_ui` (removing `inspector_ui`'s `build.rs` / `ZED_REPO_DIR`). - **corgi.toml**: declare the pinned tools (`cmake`, prebuilt libwebrtc via `LK_CUSTOM_WEBRTC`), a release-only `ZED_COMMIT_SHA` probe, and the non-`.rs` / cross-package reads each package performs. ### Validation - `corgi check` + `corgi build`, dev and release, for `zed` (plus `cli`, `remote_server`, `ep`) with a warm C++ cache — clean; built binaries embed no checkout path. - `corgi fmt --check`, `corgi clippy`, and `corgi test` on the changed crates (922 tests pass). - `cargo check -p util` locally; relying on CI for the full cargo build/test matrix. ### Notes for the reviewer - Per the repo rule, the first two lines of `README.md` are the review-confirmation marker; remove them before merging (this is what Danger is failing on). - `remote_server` cross-compilation (linux-musl via zigbuild) is not yet covered by `corgi.toml` — host (macOS arm64) only for now. ### Suggested .rules additions Only if the team wants corgi guidance in `.rules` (it recurred several times here): dev builds must not bake the checkout path into artifacts — no `env!("CARGO_MANIFEST_DIR")` or `file!()`-derived absolute paths in library/binary output. Resolve the repo root at runtime via `util::dev_repo_root()` (walks to `.git`), and use `util::dev_fs_embed!` for dev-mode `rust_embed` sources. Release Notes: - N/A --------- Co-authored-by: Ben Kunkle <ben@zed.dev>
Inspector
This is a tool for inspecting and manipulating rendered elements in Zed. It is only available in debug builds. Use the dev::ToggleInspector action to toggle inspector mode and click on UI elements to inspect them.
Current features
-
Picking of elements via the mouse, with scroll wheel to inspect occluded elements.
-
Temporary manipulation of the selected element.
-
Layout info for
Div. -
Both Rust and JSON-based style manipulation of
Divstyle. The rust style editor only supports argumentlessStyledandStyledExtmethod calls. -
Navigation to code that constructed the element.
Known bugs
JSON style editor undo history doesn't get reset
The JSON style editor appends to its undo stack on every change of the active inspected element.
I attempted to fix it by creating a new buffer and setting the buffer associated with the json_style_buffer entity. Unfortunately this doesn't work because the language server uses the version: clock::Global to figure out the changes, so would need some way to start the new buffer's text at that version.
json_style_buffer.update(cx, |json_style_buffer, cx| {
let language = json_style_buffer.language().cloned();
let file = json_style_buffer.file().cloned();
*json_style_buffer = Buffer::local("", cx);
json_style_buffer.set_language(language, cx);
if let Some(file) = file {
json_style_buffer.file_updated(file, cx);
}
});
Future features
-
Action and keybinding for entering pick mode.
-
Ability to highlight current element after it's been picked.
-
Info and manipulation of element types other than
Div. -
Indicate when the picked element has disappeared.
-
To inspect elements that disappear, it would be helpful to be able to pause the UI.
-
Hierarchy view?
Methods that take arguments in Rust style editor
Could use TreeSitter to parse out the fluent style method chain and arguments. Tricky part of this is completions - ideally the Rust Analyzer already being used by the developer's Zed would be used.
Edit original code in Rust style editor
Two approaches:
-
Open an excerpt of the original file.
-
Communicate with the Zed process that has the repo open - it would send the code for the element. This seems like a lot of work, but would be very nice for rapid development, and it would allow use of rust analyzer.
With both approaches, would need to record the buffer version and use that when referring to source locations, since editing elements can cause code layout shift.
Source location UI improvements
-
Mode to navigate to source code on every element change while picking.
-
Tracking of more source locations - currently the source location is often in a ui component. Ideally this would have a way for the components to indicate that they are probably not the source location the user is looking for.
-
Could have
InspectorElementIdbeVec<(ElementId, Option<Location>)>, but if there are multiple code paths that construct the same element this would cause them to be considered different. -
Probably better to have a separate
Vec<Option<Location>>that uses the same indices asGlobalElementId.
-
Persistent modification
Currently, element modifications disappear when picker mode is started. Handling this well is tricky. Potential features:
-
Support modifying multiple elements at once. This requires a way to specify which elements are modified - possibly wildcards in a match of the
InspectorElementIdpath. This might default to ignoring all numeric parts and just matching on the names. -
Show a list of active modifications in the UI.
-
Support for modifications being partial overrides instead of snapshots. A trickiness here is that multiple modifications may apply to the same element.
-
The code should probably distinguish the data that is provided by the element and the modifications from the inspector. Currently these are conflated in element states.
If support is added for editing original code, then the logical selector in this case would be just matches of the source path.
Code cleanups
Consider removing special side pane rendering
Currently the inspector has special rendering in the UI, but maybe it could just be a workspace item.
Pull more inspector logic out of GPUI
Currently crates/gpui/inspector.rs and crates/inspector_ui/inspector.rs are quite entangled. It seems cleaner to pull as much logic a possible out of GPUI.
Cleaner lifecycle for inspector state viewers / editors
Currently element state inspectors are just called on render. Ideally instead they would be implementors of some trait like:
trait StateInspector: Render {
fn new(cx: &mut App) -> Task<Self>;
fn element_changed(inspector_id: &InspectorElementId, window: &mut Window, cx: &mut App);
}
See div_inspector.rs - it needs to initialize itself, keep track of its own loading state, and keep track of the last inspected ID in its render function.