feat: add `cargo wdk clean` command by svasista-ms · Pull Request #638 · microsoft/windows-drivers-rs

svasista-ms · 2026-04-01T13:04:15Z

Adds clean subcommand to cargo-wdk. The command can be used to remove artifacts produced by the build command.

Closes #636

Behavior:

The command works for standalone, workspaces and emulated workspaces (with a known issue).
Standalone packages / workspace: If a Cargo.toml (manifest) is present, runs cargo clean directly. This cleans standalone packages and workspaces (excluding the directories that are mentioned in the exclude list in workspace manifest)
Emulated workspace: If Cargo.toml is not found in the current working directory, it Scans immediate subdirectories for Rust projects and cleans each one. Continues on failure and reports all errors at the end.
Known Issue: Due to the current emulated workspaces detection logic, the clean command also suffers from the same issue as #477.

Summary of changes:

CleanAction added with error types and tests
CLI integration via the clean subcommand
Integration test in build_command_test.rs that tests the behavior for standalone packages, workspaces and emulated workspaces. Replaced vanilla cargo clean with cargo wdk clean

cargo wdk clean in /examples:

cargo wdk clean in the samples repo:

clean and build commands suffer with the sample problem in nested folders inside a workspace:

Copilot

Pull request overview

Adds a new cargo wdk clean subcommand to cargo-wdk, with logic mirroring the existing “standalone/workspace vs emulated workspace” detection approach used by the build flow.

Changes:

Introduces CleanAction (plus error types and unit tests) that runs cargo clean either in the current workspace root or across immediate subdirectories (emulated workspace).
Wires the new action into the CLI as a clean subcommand.
Adds integration tests for cargo wdk clean, and updates test utilities to allow invoking the new command.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
crates/cargo-wdk/src/cli.rs	Adds `clean` subcommand and dispatches into `CleanAction`.
crates/cargo-wdk/src/actions/mod.rs	Exposes the new `clean` action module.
crates/cargo-wdk/src/actions/clean/mod.rs	Implements clean logic (workspace vs emulated workspace) and runs `cargo clean`.
crates/cargo-wdk/src/actions/clean/error.rs	Defines `CleanActionError` error types for clean flow.
crates/cargo-wdk/src/actions/clean/tests.rs	Adds unit tests for `CleanAction`.
crates/cargo-wdk/tests/clean_command_test.rs	Adds integration tests for `cargo wdk clean`.
crates/cargo-wdk/tests/test_utils/mod.rs	Allows `create_cargo_wdk_cmd` to invoke `clean`.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

…d names

codecov-commenter · 2026-04-02T09:10:39Z

Codecov Report

❌ Patch coverage is 92.28070% with 22 lines in your changes missing coverage. Please review.
✅ Project coverage is 79.45%. Comparing base (ebc705a) to head (3fd27e4).

Files with missing lines	Patch %	Lines
crates/cargo-wdk/src/actions/clean/mod.rs	94.11%	5 Missing and 10 partials ⚠️
crates/cargo-wdk/src/actions/build/mod.rs	69.23%	2 Missing and 2 partials ⚠️
crates/cargo-wdk/src/providers/fs.rs	86.66%	0 Missing and 2 partials ⚠️
crates/cargo-wdk/src/cli.rs	50.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #638      +/-   ##
==========================================
+ Coverage   78.69%   79.45%   +0.75%     
==========================================
  Files          25       26       +1     
  Lines        5234     5500     +266     
  Branches     5234     5500     +266     
==========================================
+ Hits         4119     4370     +251     
- Misses        995     1001       +6     
- Partials      120      129       +9

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

…assert removal

…ll in clean action tests

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Co-authored-by: Copilot <copilot@github.com>

Copilot

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 2 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

…icrosoft#641) This pull request updates the documentation to inform users about a new build tool in development for Rust drivers. The most important change is the addition of a preview notice about the upcoming `cargo-wdk` tool. Documentation update: * Added a preview notice in `README.md` announcing the development of the `cargo-wdk` cargo-extension, which is intended to replace `cargo-make` as the recommended tool for building Rust drivers, and encouraged users to try it out and provide feedback. --------- Signed-off-by: Krishna Kumar Thokala <krishna.thokala2010@gmail.com> Co-authored-by: Gurinder Singh <frederick.the.fool@gmail.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

…bilization (microsoft#643) Fixes `collapsable_if` lint failing in CI pipelines. The `assert_matches` stabilization that was expected to ship with Rust 1.95.0 was [reverted from the 1.95 beta](rust-lang/rust#154999) due to an [inconsistency in temporary scoping behavior](rust-lang/rust#154406) between `assert_matches!`/`assert_eq!` and their `debug_` variants. We gated `assert_matches` usage based on the nightly compiler warning that the feature was "stable since 1.95.0" (see microsoft#612). Since the stabilization was reverted before the actual release, this gate now fires on stable 1.95 and tries to use an unstable API, breaking CI pipelines. This bumps the gate to `1.96.0`. Beta 1.96 currently still has `assert_matches` available, though that may change if the destabilization propagates. --------- Co-authored-by: leon-xd <leondu@microsoft.com>

Adds and integrates `wdk::fmt::FormatBuffer`. ## Summary - Introduced `crates/wdk/src/fmt.rs` with a public `wdk::fmt` module containing two types: - `FormatBuffer<N>` — a fixed-size, stack-allocated formatting buffer implementing `fmt::Write` - `FlushableFormatBuffer<F, N>` — a wrapper that auto-flushes via closure on overflow and on drop - Swapped `print.rs` WDM/KMDF path to use `FlushableFormatBuffer` instead of `DbgPrintBufWriter`; removed that module and its tests. - Turns the `alloc` feature from the `wdk` crate into a no-op — it had no remaining uses after the stack-buffer migration. Avoids removal of feature to prevent a breaking change - Bumps MSRV from 1.85 to 1.91 due to use of [`str::char_floor_boundary`](https://doc.rust-lang.org/core/primitive.str.html#method.floor_char_boundary) ## Details - `FormatBuffer<N>` stores a zero-initialized `[u8; N]` and tracks `used` bytes. The last byte is reserved for a NUL terminator, so usable capacity is `N - 1`. `N` must be at least 2; smaller values will not compile (`const { assert!(N >= 2) }`). - All buffer mutations go through a single `append_bytes` helper that copies data and sets `buffer[used] = 0`, centralizing the NUL terminator invariant. - `as_str()` returns an infallible UTF-8 view — only valid UTF-8 enters via `write_str` (both `FormatBuffer::write_str` and `FlushableFormatBuffer::write_str` copy from `&str` sources). - `as_c_str()` returns a `&CStr` view up to the first NUL. Infallible in practice since the NUL invariant is always maintained. - `fmt::Write` clamps overflow writes at a UTF-8 char boundary via `floor_char_boundary` and signals truncation via `fmt::Error`. - `FlushableFormatBuffer<F, N>` wraps `FormatBuffer` and auto-flushes via a closure when the buffer fills. Remaining content is flushed on drop. This allows arbitrarily long formatted output to be processed in fixed-size chunks. - `print.rs` WDM/KMDF path now formats into `FlushableFormatBuffer<_, 512>` and calls `DbgPrint` via `%s` using the resulting `CStr`. - Removed the `alloc` feature gate from the WDM/KMDF print path — `print!`/`println!` no longer require heap allocation and work at any IRQL where `DbgPrint` is valid. - Custom `Debug` impl on `FormatBuffer` shows only the used content, not stale bytes after `clear()`. ## Testing 29 tests covering: - basic write, `as_str()`, `as_c_str()` usage - overflow, multi-write, and exact-fit scenarios - empty writes and min-sized buffers (`N = 2`) - clear-then-shorter-write regression (NUL terminator correctness) - multi-byte UTF-8 char boundary handling in both buffer types - explicit `flush()` mid-stream and flush-on-drop - interior NUL truncation via `as_c_str()` - `compile_fail` doctest for `N < 2` ```pwsh cargo test --package wdk --lib --all-features -- fmt --nocapture ``` ## Notes / Follow-ups - `_print` ignores `fmt::write` errors — partial output is acceptable for debug printing. Errors from individual `Display` impls are silently dropped. - The old `DbgPrintBufWriter` stripped interior NUL bytes and continued printing. The new WDM/KMDF path truncates each chunk at the first NUL (via `as_c_str`). UMDF still strips NULs and prints the remainder. --------- Signed-off-by: Leon Durrenberger <leon.durrenberger@gmail.com> Co-authored-by: leon-xd <leondu@microsoft.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

… build command to render compiler errors instead of the plain `json` option (microsoft#645) Fixes microsoft#613 ## Problem When `cargo wdk build` encounters a compilation error, the error output is hard to debug: 1. JSON wall — `--message-format=json` sends even the compiler diagnostics as JSON to `stdout` instead of redirecting it to `stderr`. So, compiler errors are not directly visible instead they are buried deep in the output json. This is affecting test outputs because in case of failures, the errors cannot be seen easily in the `json` output. 2. Redundant error content — `BuildTaskError::CargoBuild` wraps the full `CommandError` (containing `stdout` JSON), even though the user may have already seen the real compiler errors on `stderr` in real time. ## Fixes The following changes fix the above problems: 1. Switch to `--message-format=json-render-diagnostics` This option instructs cargo to render diagnostics from `rustc` directly to `stderr` instead of putting it in the `json` output. stdout: only machine-parseable JSON (artifacts, build-finished) — no diagnostic messages stderr: human-readable compiler errors and warnings, rendered by Cargo (visible to the user in real time since stderr is inherited) 2. Omit `stdout` from `CommandError::CommandFailed` before returning from `BuildTask`: When cargo build fails, `BuildTask::run()` now maps the error to a new instance of `CommandFailed` but with the `stdout` field set to `String::new()` to avoid printing machine-readable (`json`) output on the screen. ## Before Running `kmdf_driver_cross_compiles_with_cli_option_successfully` test after removing `aarch64-pc-windows-msvc` target: <img width="947" height="572" alt="image" src="https://github.com/user-attachments/assets/efb391d3-02e8-47fd-907e-91ffe5c1d4ba" /> _____ Compiler diagnostics not printed to `stderr`, instead it is buried in the json output when `cargo wdk build` fails: <img width="1547" height="822" alt="image" src="https://github.com/user-attachments/assets/4a0cf5c0-a9d4-4a8c-be63-c1237d6ed835" /> _____ Output seen when tests fail due to compilation errors: <img width="1936" height="1162" alt="image" src="https://github.com/user-attachments/assets/ac4a5e53-5a95-483d-898c-0e751e790c36" /> ## After Running `kmdf_driver_cross_compiles_with_cli_option_successfully` test after removing `aarch64-pc-windows-msvc` target: <img width="1344" height="824" alt="image" src="https://github.com/user-attachments/assets/f95f94a9-9076-46e3-a706-c07c7ca95b65" /> _____ Compiler diagnostics available on terminal: <img width="1028" height="576" alt="image" src="https://github.com/user-attachments/assets/39d6ba3f-5b6f-4fdb-bbb9-0b39c21874be" /> _____ Output seen when tests fail due to compilation errors: <img width="1867" height="1231" alt="image" src="https://github.com/user-attachments/assets/b282efcd-c748-4129-b2c1-4f5c6b297900" /> --------- Co-authored-by: Copilot <copilot@github.com>

### Summary This PR fixes three clippy lint failures in wdk-build that were [blocking CI](https://github.com/microsoft/windows-drivers-rs/actions/runs/25316776664/job/74241285479): 1. `useless_borrows_in_formatting` — removed redundant `&` from `panic!` arguments in `cargo_make.rs`. 2. `uninlined_format_args` — inlined the constant names directly into the `panic!` format strings. 3. `manual_assert_eq` — replaced `assert!(a == b)` with `assert_eq!(a, b)` in `utils.rs`. No functional changes; lint fixes only. ### Validation `cargo make wdk-pre-commit-flow` passes locally with the following `rustc` versions: 1. Stable - `rustc 1.95.0 (59807616e 2026-04-14)` 2. Nightly - `rustc 1.97.0-nightly (cb40c25f6 2026-05-04)` 3. Beta - `rustc 1.96.0-beta.5 (a5a9a5438 2026-05-01)` This PR fixes the issue. Co-authored-by: Copilot <copilot@github.com>

…tead of `ReadDirError` Co-authored-by: Copilot <copilot@github.com>

Copilot

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

svasista-ms · 2026-05-05T13:43:14Z

        #[error("Failed to rename file from {0} to {1}")]
        RenameError(PathBuf, PathBuf, #[source] io::Error),
-        #[error("Failed to get file type for directory entry {0:#?}")]
+        #[error("Failed to get file type for directory entry {0}")]


While adding this variant back, I noticed it was using {0:#?} whereas every other FileError variant uses plain {0}. Switched it to {0} so the PathBuf is rendered via Display instead of Debug to keep it consistent with the rest of the enum

krishnakumar4a4 · 2026-05-07T07:20:04Z

+                .unwrap_or_default();
+
+            if !found_at_least_one_project {
+                info!("Cleaning package(s) in {}", self.working_dir.display());


Nit: to make it more readable, Add a comment on top of this to clarify that this conditional is required to have one log per entire emulated workspace. it threw me off a bit if I didn't pay close attention to the argument being logged.

Added a comment 👍

krishnakumar4a4 · 2026-05-07T07:37:54Z

+                continue;
+            }
+
+            let working_dir_path = entry.path;


Please change this variable name, as the working_dir variable is already used to refer to the emulated workspace itself. A better name would be cargo_package_path or rust_package_path.

More generally, we may need to standardise the terminology and refer to these as Cargo projects rather than Rust projects, since we depend on Cargo.toml. I originally introduced the term Rust projects in the build flow, assuming this tool would majorly be used by first‑time Rust adopters for driver development. On reflection, however, Cargo project is the more technically accurate term.

"Package" is a cargo concept not a rustc one so cargo_package_path is better.

I renamed it to cargo_package_path 👍

krishnakumar4a4 · 2026-05-07T07:46:49Z

            }

-            let working_dir_path = dir.path(); // Avoids a short-lived temporary
+            let working_dir_path = entry.path;


This comment even applies here as we are not really switching into the directory during this loop.

https://github.com/microsoft/windows-drivers-rs/pull/638/changes#r3199700749

Renamed to cargo_package_path 👍

svasista-ms added 3 commits April 1, 2026 18:16

feat: implement clean action module with errors and tests

b9db4c5

feat: add clean subcommand to CLI

ef2d04e

feat: add integration tests for cargo wdk clean command

4075932

Copilot AI review requested due to automatic review settings April 1, 2026 13:04

Copilot started reviewing on behalf of svasista-ms April 1, 2026 13:05 View session

Copilot AI reviewed Apr 1, 2026

View reviewed changes

refactor: run clean command within build command integration tests

9ce3a8e

svasista-ms changed the title ~~Clean command~~ feat: add cargo wdk clean command Apr 2, 2026

refactor: update CleanActionError types and test names based on metho…

4b58929

…d names

svasista-ms marked this pull request as ready for review April 6, 2026 04:41

Copilot AI review requested due to automatic review settings April 6, 2026 04:41

Copilot started reviewing on behalf of svasista-ms April 6, 2026 04:42 View session

Copilot AI reviewed Apr 6, 2026

View reviewed changes

Comment thread crates/cargo-wdk/tests/build_command_test.rs Outdated

Comment thread crates/cargo-wdk/src/cli.rs

Comment thread crates/cargo-wdk/src/actions/clean/tests.rs Outdated

svasista-ms added 2 commits April 6, 2026 11:31

refactor(tests): run clean command again after the build command and …

64e7bcc

…assert removal

refactor(tests): remove unnecessary Clippy lint suppression for mocka…

fbf1c4b

…ll in clean action tests

Copilot AI review requested due to automatic review settings April 6, 2026 08:58

Copilot started reviewing on behalf of svasista-ms April 6, 2026 08:59 View session

svasista-ms requested a review from a team April 6, 2026 09:00

Copilot AI reviewed Apr 6, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/actions/clean/error.rs Outdated

Comment thread crates/cargo-wdk/src/actions/clean/tests.rs Outdated

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs Outdated

svasista-ms added 2 commits April 6, 2026 19:28

refactor: simplify project validation logic in clean action

9cbdc66

refactor: update CargoClean error

915366a

svasista-ms requested a review from Copilot April 6, 2026 14:24

Copilot started reviewing on behalf of svasista-ms April 6, 2026 14:25 View session

Copilot AI reviewed Apr 6, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs Outdated

refactor: simplify subdirectory name extraction in run method

3fe851f

svasista-ms requested a review from Copilot April 6, 2026 15:04

Copilot started reviewing on behalf of svasista-ms April 6, 2026 15:04 View session

Copilot AI reviewed Apr 6, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs

refactor: add documentation for CleanAction struct

6bb2c93

krishnakumar4a4 reviewed Apr 28, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs Outdated

refactor: move tests into CleanAction mod.rs file

80cc4ec

svasista-ms mentioned this pull request Apr 28, 2026

[cargo-wdk] Gate mockall/mockall_double behind cfg(test), move to [dev-dependencies], and clean up the provider/consumer mocking pattern #651

Open

krishnakumar4a4 previously approved these changes Apr 28, 2026

View reviewed changes

gurry reviewed Apr 30, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/actions/clean/mod.rs Outdated

refactor: directory entry handling in CleanAction and BuildAction

abf8412

Co-authored-by: Copilot <copilot@github.com>

Copilot AI review requested due to automatic review settings May 4, 2026 13:47

svasista-ms dismissed krishnakumar4a4’s stale review via abf8412 May 4, 2026 13:47

Copilot started reviewing on behalf of svasista-ms May 4, 2026 13:48 View session

Copilot AI reviewed May 4, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/providers/fs.rs Outdated

Comment thread crates/cargo-wdk/src/actions/build/mod.rs Outdated

gurry reviewed May 5, 2026

View reviewed changes

Comment thread crates/cargo-wdk/src/cli.rs Outdated

krishnakumar4a4 and others added 8 commits May 5, 2026 18:46

fix: correct typo in debug log message

42a67e3

refactor: use DirFileTypeError variant when file_type() fails ins…

b23b0b7

…tead of `ReadDirError` Co-authored-by: Copilot <copilot@github.com>

fix: about string for the clean subcommand

90a8726

Copilot AI review requested due to automatic review settings May 5, 2026 13:29

Copilot started reviewing on behalf of svasista-ms May 5, 2026 13:31 View session

Merge branch 'main' into clean-command

69f895e

Copilot AI reviewed May 5, 2026

View reviewed changes

svasista-ms commented May 5, 2026

View reviewed changes

gurry previously approved these changes May 6, 2026

View reviewed changes

gurry enabled auto-merge May 6, 2026 03:49

svasista-ms requested a review from krishnakumar4a4 May 7, 2026 03:37

krishnakumar4a4 requested changes May 7, 2026

View reviewed changes

refactor: update variable names for clarity in build and clean actions

3fd27e4

svasista-ms dismissed gurry’s stale review via 3fd27e4 May 7, 2026 18:07

Conversation

svasista-ms commented Apr 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Behavior:

Summary of changes:

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

codecov-commenter commented Apr 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

svasista-ms May 5, 2026

Choose a reason for hiding this comment

Uh oh!

krishnakumar4a4 May 7, 2026

Choose a reason for hiding this comment

Uh oh!

svasista-ms May 8, 2026

Choose a reason for hiding this comment

Uh oh!

krishnakumar4a4 May 7, 2026

Choose a reason for hiding this comment

Uh oh!

gurry May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

svasista-ms May 8, 2026

Choose a reason for hiding this comment

Uh oh!

krishnakumar4a4 May 7, 2026

Choose a reason for hiding this comment

svasista-ms commented Apr 1, 2026 •

edited

Loading

codecov-commenter commented Apr 2, 2026 •

edited

Loading

gurry May 8, 2026 •

edited

Loading