refactor: drop Zig-shadowing underscore affixes from Rust fn names - #36067
Merged
Jarred-Sumner merged 5 commits intoJul 27, 2026
Merged
Conversation
In the original Zig codebase many functions were named with a leading or
trailing underscore to work around Zig compiler restrictions on variable
shadowing (e.g. `resume_` because `resume` is a Zig keyword, `_pass`
because another function had a local `pass` variable). Rust has no such
restriction, so these names can be cleaned up.
This renames ~55 functions across 42 files where the clean name does not
collide with anything in the same scope:
Trailing underscore dropped (Zig keyword or shadowing holdover):
resume_ is_function_ or_ jsdom_file_construct_ on_handshake_ (valkey/mysql)
server_set_{idle_timeout,on_client_error,on_connection,app_flags,max_http_header_size}_
RusageFields::{maxrss,ixrss,nswap,inblock,oublock,msgsnd,msgrcv,nsignals,nvcsw,nivcsw}_
SQLDataCell::bool_ Frame/Reader::u32_
Leading underscore dropped (Zig private-impl convention):
_pass (expect), _advance _parse (dotenv) _resolve (jsc_hooks)
_boot_and_handle_error _stat _socket _decode _format_t _count_with_hash
_buf_append_input _join_abs_string_buf_windows _on_read_chunk
_on_structured_clone_deserialize _compare _compare_ipv6 _generic_flush
_generic_write _events_cb _schedule _stop _cp_async_directory _cp_symlink
_cp_open_dest_with_mkdir _copy_single_file_sync
ipc: _socket_closed _write _on_write_complete _on_after_ipc_closed
_close_socket_task _windows_close _windows_on_closed _windows_on_write_complete
Not renamed (need the underscore in Rust too):
ref_ loop_ final_ mut_ do_ typeof_ static_ match_ as_ use_ impl_ for_
(Rust keywords/reserved)
deflateInit_ inflateInit_ gzgetc_ _dyld_* _lwp_self _strnicmp _umask
(real C symbol names)
on_/get_attribute_/hash_/digest_/cast_/get_/fast_*_/start_/next_/ptr_/etc.
(same impl block already has the clean name as a wrapper)
_atomic_* _dispatch_* _ws_minify* _wired _needs_nl
(intentional Rust doc-hidden / macro helpers)
Contributor
|
Warning Review limit reachedYou’ve reached a temporary PR review limit under our Fair Usage Limits Policy. Next review available in: 3 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (37)
WalkthroughChangesThis PR removes trailing underscores from numerous public and internal Rust helpers, callbacks, accessors, and codec methods. Call sites, FFI shims, comments, and JavaScript host mappings are updated while preserving the described behavior. Underscore-free API and data contracts
Runtime and filesystem integration
Possibly related PRs
🚥 Pre-merge checks | ✅ 4✅ Passed checks (4 passed)
Comment |
Collaborator
Author
|
No new tests: this is a mechanical rename refactor with zero behavioral change (every renamed function has identical signature and body; call sites are updated 1:1). There is no test that would fail on |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/runtime/cli/run_command.rs`:
- Around line 1744-1746: Update the doc comment above boot_and_handle_error to
reference the current helper name instead of the obsolete _bootAndHandleError
identifier, leaving the described behavior unchanged.
In `@src/runtime/node/fs_events.rs`:
- Line 527: Update the lifecycle documentation associated with events_cb and
schedule in path_watcher.rs, replacing the outdated _events_cb and _schedule
references with the current symbol names. Keep the teardown contract description
accurate and limited to durable, non-obvious information.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 437e2802-46c3-435c-b721-eb278ea4a7f0
📒 Files selected for processing (42)
src/css/lib.rssrc/css/selectors/parser.rssrc/css/selectors/selector.rssrc/dotenv/env_loader.rssrc/install/lockfile.rssrc/io/PipeReader.rssrc/jsc/NodeModuleModule.rssrc/jsc/ipc.rssrc/paths/Path.rssrc/paths/resolve_path.rssrc/resolver/data_url.rssrc/runtime/api/bun/h2_frame_parser.rssrc/runtime/api/bun/subprocess/ResourceUsage.rssrc/runtime/api/standalone_graph_jsc.rssrc/runtime/cli/run_command.rssrc/runtime/cli/test/parallel/Coordinator.rssrc/runtime/cli/test/parallel/Frame.rssrc/runtime/cli/test/parallel/Worker.rssrc/runtime/cli/test/parallel/runner.rssrc/runtime/hw_exports.rssrc/runtime/jsc_hooks.rssrc/runtime/node/fs_events.rssrc/runtime/node/net/BlockList.rssrc/runtime/node/node_fs.rssrc/runtime/node/path.rssrc/runtime/server/NodeHTTPResponse.rssrc/runtime/server/mod.rssrc/runtime/server/server_body.rssrc/runtime/test_runner/bun_test.rssrc/runtime/test_runner/expect.rssrc/runtime/test_runner/jest.classes.tssrc/runtime/valkey_jsc/js_valkey.rssrc/runtime/webcore/Blob.rssrc/spawn_sys/spawn_process.rssrc/sql_jsc/mysql/JSMySQLConnection.rssrc/sql_jsc/mysql/protocol/DecodeBinaryValue.rssrc/sql_jsc/mysql/protocol/ResultSet.rssrc/sql_jsc/postgres/DataCell.rssrc/sql_jsc/postgres/PostgresSQLConnection.rssrc/sql_jsc/shared/SQLDataCell.rssrc/uws_sys/Response.rssrc/uws_sys/h3.rs
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
Follow-up to the function renames: struct fields named with a trailing
underscore because the clean name is a Zig keyword (or shadowed a local
in Zig). None of these are Rust keywords.
Fields renamed:
test_ -> test (AST If/For/DoWhile/While/Switch, e::If; ~105 sites)
catch_ -> catch (AST Try; ~11 sites)
cache_directory_ -> cache_directory (PackageManager)
error_ -> error (NativeBrotli Context)
event_type_ -> event_type (ChangeEvent)
file_polls_ -> file_polls (RareData, MiniEventLoop)
process_ -> process (WindowsSpawnResult)
Not renamed (kept with underscore):
Rust keywords: type_ loop_ ref_ impl_ macro_ fn_ in_ static_ do_ self_ move_
Field+accessor-method pairs (clean name is the accessor): Entry::base_,
PackedMap::vlq_, RareData::{mysql,valkey,ws_*}_group_, spawn_sync_event_loop_
repr(C) mirrors of C struct fields that carry the underscore in C:
libuv sys_errno_/unused_, libwebp private_, ExternSocketConfig::unix_
(C macro), SQLDataCell Value::bool_
cares AddrInfo::name_/cnames_ (fn name() accessor uses the clean name)
Also from review:
- revert NthSelectorData::is_function_ (field with same name exists and
answers a different question)
- rename postgres SocketHandler::on_handshake_ to match valkey/mysql
(delegates to a different type, no collision)
- delete dead NodeFS::_cp_async_directory wrapper
- fix stale doc-comment references to renamed fns
The _resolve -> resolve sed in jsc_hooks.rs over-matched two doc comments that were cross-referencing VirtualMachine::_resolve (which this PR did not rename, as VirtualMachine already has a separate public resolve()). Restore the underscore so the comments point at the right method.
Contributor
There was a problem hiding this comment.
LGTM — all prior review findings addressed; mechanical rename with compiler-enforced call-site coverage.
What was reviewed:
- Verified all five earlier findings are fixed on HEAD:
path_watcher.rs_events_cbrefs updated,NthSelectorData::is_function_revert kept, deadNodeFS::_cp_async_directoryremoved, postgreson_handshakerenamed,VirtualMachine._resolvecomment refs restored at jsc_hooks.rs:4807/4850. - Spot-checked the
jest.classes.tsfn: "pass"binding matches the renamedExpect::pass. - Grepped for lingering
.test_/.catch_AST field accesses — none remain. - Confirmed the deleted
#[cfg(not(windows))] _windows_closestub had no callers (Windows path uses the#[cfg(windows)]variant).
Extended reasoning...
Overview
Pure identifier-rename refactor across 71 files: ~60 functions and ~12 struct fields drop leading/trailing underscores that existed only as Zig keyword/shadowing workarounds (test_ → test, catch_ → catch, resume_ → resume, _resolve → resolve, etc.). Bodies and signatures are byte-identical; call sites updated 1:1. One codegen input (jest.classes.ts) updated to match Expect::_pass → pass. Two dead-code items removed (unused #[cfg(not(windows))] _windows_close stub, uncalled NodeFS::_cp_async_directory wrapper). Two #[cfg_attr(..., allow(dead_code))] attributes added where the _ prefix was previously suppressing the lint on platform-conditionally-unused fns.
Security risks
None. No user-facing surface changes, no parsing/validation logic touched, no auth/crypto/permissions paths affected. The renamed identifiers are internal Rust symbols; the one .classes.ts change maps a JS prototype method name to its (renamed) Rust impl — the JS-visible name pass is unchanged.
Level of scrutiny
Low-to-medium. The compiler is the primary reviewer for a rename refactor: bun run rust:check-all passes on all six targets (linux/macos/windows × x64/aarch64), which proves every call site was updated and no #[cfg]-gated path was missed. The residual risk was (a) comment drift, (b) accidental field/method name collisions, and (c) missed sibling sites — all three were caught in earlier review rounds and fixed in c72ccc4 / cd5ed70. This run's bug-hunting pass found nothing new.
Other factors
Two prior review rounds surfaced five nits (stale cross-file comment refs, an is_function field/method collision, a dead wrapper, a missed postgres sibling, and a sed over-match on VirtualMachine._resolve); the author addressed each and I verified the fixes on HEAD via targeted greps. All inline threads are resolved. The PR description's "Not renamed" section documents the deliberate exclusions (Rust keywords, C ABI names, field+accessor pairs), and the one item that violated its own exclusion rule (NthSelectorData::is_function_) was reverted. Existing test coverage (expect.test.js, fs/cp.test.ts, transpiler.test, bundler tests) exercises the renamed paths.
This was referenced Jul 27, 2026
Jarred-Sumner
pushed a commit
that referenced
this pull request
Jul 27, 2026
Main fails to build on all lanes since bfd1e92 (Buildkite main #83211, #83215). ## Cause #36067 renamed `AnyResponse::resume_` to `AnyResponse::resume` in `src/uws_sys/Response.rs`. #36006, merged just before it, added six new callers of the old name. Both passed CI on their own branches; combined on main: ``` error[E0599]: no method named `resume_` found for enum `AnyResponse` in the current scope --> src/runtime/server/RequestContext.rs:2382:22 --> src/runtime/server/RequestContext.rs:4126:18 --> src/runtime/server/RequestContext.rs:4182:22 --> src/runtime/webcore/streams.rs:1408:25 --> src/runtime/webcore/streams.rs:1785:17 --> src/runtime/webcore/streams.rs:1873:21 ``` ## Fix Rename the six call sites to `resume()`. ## Verification - `cargo check -p bun_runtime` on main: 6× E0599; with this change: clean. - `bun bd` builds. - `bun bd test test/js/bun/http/serve.test.ts -t 'request body backpressure'`: 5/5 pass (the tests #36006 added that exercise these call sites). <!-- robobun:evidence:begin --> --- **no test proof** · iteration 1 · Platform-specific test(s) that do not run on this machine. Deferring to CI, which covers all platforms: test/js/bun/http/serve.test.ts <!-- robobun:evidence:end -->
Jarred-Sumner
pushed a commit
that referenced
this pull request
Jul 27, 2026
## What Adds `src/runtime/node/node_fs.rs: 2` to the dead-code escape inventory so the `dead-code-escapes` source-lint passes again on main. ## Why #36067 (bfd1e92) renamed `_cp_symlink` → `cp_symlink` and `_cp_open_dest_with_mkdir` → `cp_open_dest_with_mkdir`, moving the dead-code suppression from the leading underscore (which the inventory regex does not match) to explicit `#[cfg_attr(..., allow(dead_code))]` attributes: ```rust #[cfg_attr(any(windows, target_os = "macos"), allow(dead_code))] fn cp_symlink(&mut self, ...) { ... } #[cfg_attr(windows, allow(dead_code))] fn cp_open_dest_with_mkdir(&mut self, ...) { ... } ``` Both functions are called from the Linux/FreeBSD arms of `copy_single_file_sync` (lines 8613, 8680, 8706, 8852, 8878) and are only dead on the platforms named in the predicate, so the escapes are legitimate per the test's own rules. The inventory file was simply not regenerated, and the GitHub Actions `source-lints` workflow has been failing on every main commit since: ``` (fail) #[allow(dead_code)] escapes > src/runtime/node/node_fs.rs (0) error: src/runtime/node/node_fs.rs has 2 item-level #[allow(dead_code)] escapes, up from 0. ``` Failed runs: 30249533282 (bfd1e92), 30249969141, 30250537591, 30250867335. ## How Regenerated with `bun ./test/internal/source-lints/dead-code-escapes.test.ts` per the test's instructions. ## Verification ``` $ bun test test/internal/source-lints/dead-code-escapes.test.ts ... (pass) #[allow(dead_code)] escapes > src/runtime/node/node_fs.rs (2) ... 26 pass 0 fail ``` <!-- robobun:evidence:begin --> --- **no test proof** · iteration 0 · docs-only change; test-proof not applicable <!-- robobun:evidence:end -->
Jarred-Sumner
pushed a commit
that referenced
this pull request
Jul 27, 2026
…36067) ## What Renames ~60 functions and ~12 struct fields across 75 files to drop leading/trailing underscores that were carried over from the Zig codebase as shadowing workarounds and are not needed in Rust. Example: `resume_` is named that way because `resume` is a Zig keyword. It is not a Rust keyword, so `Response::resume()` is fine. Example: `Expect::_pass` was named that way because other functions in the file had a local `pass` variable, which Zig refuses to shadow. Rust doesn't care, so it's now `Expect::pass` (jest.classes.ts updated to match). Example: AST node fields `test_`/`catch_` exist because `test` and `catch` are Zig keywords. Neither is a Rust keyword. ## Functions renamed **Trailing underscore dropped** (was a Zig keyword or shadowed a local): - `resume_` (uws Response/h3) - `or_` (VendorPrefix; `or` is a Zig keyword, not Rust) - `jsdom_file_construct_` - `on_handshake_` (valkey, mysql, postgres SocketHandler) - `server_set_{idle_timeout,on_client_error,on_connection,app_flags,max_http_header_size}_` - `RusageFields::{maxrss,ixrss,nswap,inblock,oublock,msgsnd,msgrcv,nsignals,nvcsw,nivcsw}_` - `SQLDataCell::bool_`, `Frame::u32_`, `Reader::u32_` **Leading underscore dropped** (Zig private-impl convention): - `Expect::_pass` (+ jest.classes.ts) - `_advance` `_parse` (dotenv) `_resolve` (jsc_hooks) `_boot_and_handle_error` `_stat` `_socket` `_decode` `_format_t` `_count_with_hash` `_buf_append_input` `_join_abs_string_buf_windows` `_on_read_chunk` `_on_structured_clone_deserialize` `_compare` `_compare_ipv6` `_generic_flush` `_generic_write` `_events_cb` `_schedule` `_stop` - node_fs: `_cp_async_directory` `_cp_symlink` `_cp_open_dest_with_mkdir` `_copy_single_file_sync` - ipc: `_socket_closed` `_write` `_on_write_complete` `_on_after_ipc_closed` `_close_socket_task` `_windows_close` `_windows_on_closed` `_windows_on_write_complete` For `cp_symlink`/`cp_open_dest_with_mkdir` (only called on some platforms) the dead-code suppression moves from the `_` prefix to an explicit `#[cfg_attr(..., allow(dead_code))]`. The unused `#[cfg(not(windows))]` `_windows_close` stub in ipc.rs and the dead `NodeFS::_cp_async_directory` wrapper are removed. ## Struct fields renamed - `test_` -> `test` (AST `If`/`For`/`DoWhile`/`While`/`Switch`, `e::If`; ~105 access sites) - `catch_` -> `catch` (AST `Try`) - `cache_directory_` (PackageManager), `error_` (NativeBrotli Context), `event_type_` (ChangeEvent), `file_polls_` (RareData, MiniEventLoop), `process_` (WindowsSpawnResult) ## Not renamed - **Rust keywords/reserved**: `ref_` `loop_` `type_` `final_` `mut_` `do_` `typeof_` `static_` `match_` `as_` `use_` `impl_` `for_` `fn_` `in_` `self_` `move_` `macro_`. These still need the suffix. - **Real C symbol names**: `deflateInit_` `inflateInit2_` `gzgetc_` `_dyld_*` `_lwp_self` `_strnicmp` `_umask` etc. - **repr(C) fields matching the C struct**: libuv `sys_errno_`/`unused_`, libwebp `private_`, `ExternSocketConfig::unix_` (C macro collision), `SQLDataCell::Value::bool_`. - **Same-scope wrapper/impl pairs**: html_rewriter `on_`/`get_attribute_`/..., CryptoHasher `hash_`/`digest_`, FetchHeaders `get_`/`cast_`/`fast_*_`, `NthSelectorData::is_function_`, `start_`/`next_`/`ptr_`/`generate_js_renamer_`/`_parse` (js_parser)/`_exec`/`_resolve` (VM)/etc. In each case the clean name already exists on the same type as a field, arg-decode wrapper, or public entrypoint, so the `_` variant still needs disambiguation. Happy to follow up with an `_impl` suffix convention if preferred. - **Field+accessor pairs** (clean name is the getter): `Entry::base_`, `PackedMap::vlq_`, `RareData::{mysql,valkey,ws_*}_group_`, `spawn_sync_event_loop_`, c-ares `AddrInfo::name_`. - **Intentional Rust `#[doc(hidden)]`/macro helpers**: `_atomic_*` `_dispatch_*` `_ws_minify*` `_wired` `_needs_nl` `_scoped_use_ansi`. - `deref_`: kept for symmetry with `ref_`. ## Verification `bun run rust:check-all` passes (linux/macos/windows, x64/aarch64). Smoke-tested `expect.test.js`, `html-rewriter.test`, `fs/cp.test.ts`, `blocklist-gc.test.ts`, `transpiler.test`, `bundler/esbuild/default.test.ts`. --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Jarred-Sumner
pushed a commit
that referenced
this pull request
Jul 27, 2026
Main fails to build on all lanes since bfd1e92 (Buildkite main #83211, #83215). ## Cause #36067 renamed `AnyResponse::resume_` to `AnyResponse::resume` in `src/uws_sys/Response.rs`. #36006, merged just before it, added six new callers of the old name. Both passed CI on their own branches; combined on main: ``` error[E0599]: no method named `resume_` found for enum `AnyResponse` in the current scope --> src/runtime/server/RequestContext.rs:2382:22 --> src/runtime/server/RequestContext.rs:4126:18 --> src/runtime/server/RequestContext.rs:4182:22 --> src/runtime/webcore/streams.rs:1408:25 --> src/runtime/webcore/streams.rs:1785:17 --> src/runtime/webcore/streams.rs:1873:21 ``` ## Fix Rename the six call sites to `resume()`. ## Verification - `cargo check -p bun_runtime` on main: 6× E0599; with this change: clean. - `bun bd` builds. - `bun bd test test/js/bun/http/serve.test.ts -t 'request body backpressure'`: 5/5 pass (the tests #36006 added that exercise these call sites). <!-- robobun:evidence:begin --> --- **no test proof** · iteration 1 · Platform-specific test(s) that do not run on this machine. Deferring to CI, which covers all platforms: test/js/bun/http/serve.test.ts <!-- robobun:evidence:end -->
Jarred-Sumner
pushed a commit
that referenced
this pull request
Jul 27, 2026
## What Adds `src/runtime/node/node_fs.rs: 2` to the dead-code escape inventory so the `dead-code-escapes` source-lint passes again on main. ## Why #36067 (bfd1e92) renamed `_cp_symlink` → `cp_symlink` and `_cp_open_dest_with_mkdir` → `cp_open_dest_with_mkdir`, moving the dead-code suppression from the leading underscore (which the inventory regex does not match) to explicit `#[cfg_attr(..., allow(dead_code))]` attributes: ```rust #[cfg_attr(any(windows, target_os = "macos"), allow(dead_code))] fn cp_symlink(&mut self, ...) { ... } #[cfg_attr(windows, allow(dead_code))] fn cp_open_dest_with_mkdir(&mut self, ...) { ... } ``` Both functions are called from the Linux/FreeBSD arms of `copy_single_file_sync` (lines 8613, 8680, 8706, 8852, 8878) and are only dead on the platforms named in the predicate, so the escapes are legitimate per the test's own rules. The inventory file was simply not regenerated, and the GitHub Actions `source-lints` workflow has been failing on every main commit since: ``` (fail) #[allow(dead_code)] escapes > src/runtime/node/node_fs.rs (0) error: src/runtime/node/node_fs.rs has 2 item-level #[allow(dead_code)] escapes, up from 0. ``` Failed runs: 30249533282 (bfd1e92), 30249969141, 30250537591, 30250867335. ## How Regenerated with `bun ./test/internal/source-lints/dead-code-escapes.test.ts` per the test's instructions. ## Verification ``` $ bun test test/internal/source-lints/dead-code-escapes.test.ts ... (pass) #[allow(dead_code)] escapes > src/runtime/node/node_fs.rs (2) ... 26 pass 0 fail ``` <!-- robobun:evidence:begin --> --- **no test proof** · iteration 0 · docs-only change; test-proof not applicable <!-- robobun:evidence:end -->
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Renames ~60 functions and ~12 struct fields across 75 files to drop leading/trailing underscores that were carried over from the Zig codebase as shadowing workarounds and are not needed in Rust.
Example:
resume_is named that way becauseresumeis a Zig keyword. It is not a Rust keyword, soResponse::resume()is fine.Example:
Expect::_passwas named that way because other functions in the file had a localpassvariable, which Zig refuses to shadow. Rust doesn't care, so it's nowExpect::pass(jest.classes.ts updated to match).Example: AST node fields
test_/catch_exist becausetestandcatchare Zig keywords. Neither is a Rust keyword.Functions renamed
Trailing underscore dropped (was a Zig keyword or shadowed a local):
resume_(uws Response/h3)or_(VendorPrefix;oris a Zig keyword, not Rust)jsdom_file_construct_on_handshake_(valkey, mysql, postgres SocketHandler)server_set_{idle_timeout,on_client_error,on_connection,app_flags,max_http_header_size}_RusageFields::{maxrss,ixrss,nswap,inblock,oublock,msgsnd,msgrcv,nsignals,nvcsw,nivcsw}_SQLDataCell::bool_,Frame::u32_,Reader::u32_Leading underscore dropped (Zig private-impl convention):
Expect::_pass(+ jest.classes.ts)_advance_parse(dotenv)_resolve(jsc_hooks)_boot_and_handle_error_stat_socket_decode_format_t_count_with_hash_buf_append_input_join_abs_string_buf_windows_on_read_chunk_on_structured_clone_deserialize_compare_compare_ipv6_generic_flush_generic_write_events_cb_schedule_stop_cp_async_directory_cp_symlink_cp_open_dest_with_mkdir_copy_single_file_sync_socket_closed_write_on_write_complete_on_after_ipc_closed_close_socket_task_windows_close_windows_on_closed_windows_on_write_completeFor
cp_symlink/cp_open_dest_with_mkdir(only called on some platforms) the dead-code suppression moves from the_prefix to an explicit#[cfg_attr(..., allow(dead_code))]. The unused#[cfg(not(windows))]_windows_closestub in ipc.rs and the deadNodeFS::_cp_async_directorywrapper are removed.Struct fields renamed
test_->test(ASTIf/For/DoWhile/While/Switch,e::If; ~105 access sites)catch_->catch(ASTTry)cache_directory_(PackageManager),error_(NativeBrotli Context),event_type_(ChangeEvent),file_polls_(RareData, MiniEventLoop),process_(WindowsSpawnResult)Not renamed
ref_loop_type_final_mut_do_typeof_static_match_as_use_impl_for_fn_in_self_move_macro_. These still need the suffix.deflateInit_inflateInit2_gzgetc__dyld_*_lwp_self_strnicmp_umasketc.sys_errno_/unused_, libwebpprivate_,ExternSocketConfig::unix_(C macro collision),SQLDataCell::Value::bool_.on_/get_attribute_/..., CryptoHasherhash_/digest_, FetchHeadersget_/cast_/fast_*_,NthSelectorData::is_function_,start_/next_/ptr_/generate_js_renamer_/_parse(js_parser)/_exec/_resolve(VM)/etc. In each case the clean name already exists on the same type as a field, arg-decode wrapper, or public entrypoint, so the_variant still needs disambiguation. Happy to follow up with an_implsuffix convention if preferred.Entry::base_,PackedMap::vlq_,RareData::{mysql,valkey,ws_*}_group_,spawn_sync_event_loop_, c-aresAddrInfo::name_.#[doc(hidden)]/macro helpers:_atomic_*_dispatch_*_ws_minify*_wired_needs_nl_scoped_use_ansi.deref_: kept for symmetry withref_.Verification
bun run rust:check-allpasses (linux/macos/windows, x64/aarch64). Smoke-testedexpect.test.js,html-rewriter.test,fs/cp.test.ts,blocklist-gc.test.ts,transpiler.test,bundler/esbuild/default.test.ts.