http_types: delete unused can_open_in_editor and URLPath::is_root/path_without_asset_prefix

[dylan-conway]

Dead-code sweep of bun_http_types.

cargo check --workspace passes.

Why this code exists

Item	Zig original	Zig callers	Status
MimeType::can_open_in_editor	src/http_types/MimeType.zig:64 (canOpenInEditor)	0	dead in Zig too
URLPath::is_root	src/http_types/URLPath.zig:13 (isRoot)	0	dead in Zig too
URLPath::path_without_asset_prefix	URLPath.zig:21 (pathWithoutAssetPrefix)	0 — only caller is isRoot at :14 (same file), itself dead	dead in Zig too

[coderabbitai]

Warning

Review limit reached

@dylan-conway, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 11 minutes. Learn how PR review limits work.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro

Run ID: b6ab33bd-698d-4138-81cc-88f46675fa72

📥 Commits

Reviewing files that changed from the base of the PR and between eae19b3 and f730821.

📒 Files selected for processing (2)

• src/CLAUDE.md
• src/http_types/MimeType.rs

Walkthrough

PR removes MimeType's can_open_in_editor() method and refactors URLPath's public API by removing is_root() and path_without_asset_prefix() methods while adding take_decoded_storage() for explicit storage ownership transfer.

Changes

HTTP types API refactoring

Layer / File(s)	Summary
MimeType editor capability removal src/http_types/MimeType.rs	Removes the can_open_in_editor() method that previously returned true for text/code MIME categories and SVG images only.
URLPath storage ownership refactoring src/http_types/URLPath.rs	Removes is_root() and path_without_asset_prefix() helper methods and adds take_decoded_storage() to transfer ownership of the decoded storage buffer to the caller, marked with #[must_use] to prevent slice-dangling bugs.

🚥 Pre-merge checks | ✅ 3 | ❌ 1

❌ Failed checks (1 inconclusive)

Check name	Status	Explanation	Resolution
Description check	❓ Inconclusive	The description addresses the required template sections with clear context, but is missing a dedicated 'What does this PR do?' section heading and verification methodology.	Add explicit 'What does this PR do?' and 'How did you verify your code works?' section headings to match the repository template structure.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the main change: removal of three unused methods from the http_types module.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[robobun]

^{Updated 1:05 AM PT - May 29th, 2026}

❌ @dylan-conway, your commit f730821 has 2 failures in Build #58870 (All Failures):

• test/js/bun/terminal/terminal.test.ts - code 1 on 🍎 14 x64
• test/js/bun/http/bun-serve-html.test.ts - pid 5527 segmentation fault on 🐧 25.04 x64

---

🧪   To try this PR locally:

bunx bun-pr 31552

That installs a local version of the PR into your bun-31552 executable, so you can run:

bun-31552 --bun

[claude]

LGTM — thanks for folding in the is_code() cascade; the remaining is_source_map field nit is non-blocking.

Extended reasoning...

Overview

Pure dead-code sweep in bun_http_types: removes MimeType::can_open_in_editor(), Category::is_code() (added in the follow-up commit after my earlier note), URLPath::is_root(), and URLPath::path_without_asset_prefix(), plus a one-line doc example swap in src/CLAUDE.md. Net change is ~50 lines deleted, 1 line modified. I verified via grep that no Rust callers remain for any of the removed items.

Security risks

None. This is deletion-only; no new code paths, inputs, or behavior. The deleted functions were unreachable helpers with no auth/crypto/IO surface.

Level of scrutiny

Low. Removing provably-uncalled pub methods from an internal crate carries effectively zero runtime risk — cargo check --workspace passing is the right verification bar, and the PR description documents the Zig-side dead-code analysis for each item. The CodeRabbit walkthrough mistakenly describes take_decoded_storage() as new, but the diff shows it as unchanged context only.

Other factors

My previous inline comment about the is_code() cascade was addressed in commit f730821. The new inline nit (the now-write-only URLPath::is_source_map struct field) is the same one-hop cascade pattern on the URLPath side — harmless to leave, fine to fold in if you want a complete sweep. The musl LTO build failure reported by robobun on the earlier commit ("Linking two modules of different data layouts … '' vs …") is an llvm-link/bitcode infrastructure issue unrelated to deleting these Rust methods.

[claude]

🟡 🟡 Same cascade pattern on the URLPath side: deleting path_without_asset_prefix() orphans the pub is_source_map: bool struct field (line 18) — its sole reader was the deleted if self.is_source_map && out.ends_with(b".map"), and the Zig field's only reader is inside pathWithoutAssetPrefix (documented dead-in-Zig in this PR's table). The field decl and its struct-literal write at line ~176 could go in the same sweep. Note: the local is_source_map in parse() (lines 146/148/171) is still live — only the struct field is dead.

Extended reasoning...

What this is

Not a correctness bug — an incomplete sweep, the same one-hop-downstream cascade already addressed for Category::is_code() in commit f730821. Deleting URLPath::path_without_asset_prefix() removes the only reader of the pub is_source_map: bool struct field, so the field now meets the identical "0 Rust readers / dead in Zig too" criterion this PR uses for the other deletions.

Step-by-step proof that the struct field is now dead

1. Rust side, before this PR: the only read of self.is_source_map was inside path_without_asset_prefix() at the line being deleted:

   if self.is_source_map && out.ends_with(b".map") {
       return &out[0..out.len() - 4];
   }

2. Rust side, after this PR: grepping all *.rs for is_source_map yields only URLPath.rs hits — the field declaration at :18, the local variable in parse() at :146/:148/:171, and the struct-literal write at :176. None of the six Rust files that reference URLPath (fetch.rs, filesystem_router.rs, router/lib.rs, http_types/lib.rs, http/lib.rs) touch the field. Zero field readers remain.
3. Zig side: the field is declared at URLPath.zig:11 and written at :135. Its only read is URLPath.zig:28, inside pathWithoutAssetPrefix — which the PR description's own table documents as having 0 callers (its only caller isRoot is itself dead). So the field is transitively dead in Zig too.
4. Conclusion: URLPath::is_source_map meets the same "0 Rust readers / dead in Zig too" bar that justified deleting is_root() / path_without_asset_prefix() / can_open_in_editor(), and is the exact analogue of the is_code() cascade already accepted and fixed.

What is not dead

The local let is_source_map = extname == b"map"; at parse():146 is still live — it drives the extname/path recomputation at :148 and :171. Only the struct field (line 18) and its assignment in the struct literal (line 176, plus the doc comment at 16–17) are dead.

Why nothing else flags this

The field is pub, so rustc's dead-code lint won't fire. There's no test or external consumer reading it. It survives only because it's one hop downstream of the function being deleted, and the sweep stopped at the first level — same as is_code() was before the follow-up commit.

Impact

None on runtime behavior — purely about completing the cleanup. Leaving the field in place is harmless; a later sweep would catch it.

Suggested fix

Either:

• Extend this PR to also delete pub is_source_map: bool (URLPath.rs:16-18) and the is_source_map, line in the struct literal (:176) — the local variable stays, or
• Intentionally keep it as parser output for Zig-port parity — in which case feel free to disregard.