[dlunch]

Changes

• Centralize UTF-16-based modified UTF-8 encoding and decoding in jvm-types, using no_std and alloc.
• Reuse the codec in classfile parsing and DataInputStream.readUTF / DataOutputStream.writeUTF, preserving stream exceptions and encoded-length limits.
• Reject standard four-byte UTF-8 in classfile constants instead of retaining the previous compatibility path.
• Extend tests for exact encoded bytes, unpaired surrogates, malformed sequences, and the 65,535-byte output limit.

Verification

• cargo test --workspace passed (two existing ignored tests).
• cargo clippy --workspace --all-targets -- -D warnings passed.
• cargo fmt --all -- --check passed.
• cargo check -p jvm-types -p classfile -p rustjava-runtime --lib --target wasm32-unknown-unknown passed.
• Full-workspace WebAssembly checking is blocked by the existing test-utils Tokio rt-multi-thread feature.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-14T04:05:57.971687Z	c8cc5d4	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.20%. Comparing base (a83ced6) to head (c8cc5d4).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #214   +/-   ##
=======================================
  Coverage   87.19%   87.20%           
=======================================
  Files         308      309    +1     
  Lines       46374    46332   -42     
=======================================
- Hits        40437    40403   -34     
+ Misses       5937     5929    -8     

☔ View full report in Codecov by Harness.
📢 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]

🔵 Needs a closer look

It changes the core class-file constant-pool parsing path and intentionally drops a previously-accepted encoding (standard 4-byte UTF-8), so a human should confirm that acceptance-behavior change is acceptable.

0 open findings

What changed in this PR

This PR removes three separate, hand-rolled modified UTF-8 implementations and replaces them with a single shared codec in the jvm-types crate. Previously, classfile constant-pool parsing, DataInputStream.readUTF, and DataOutputStream.writeUTF each had their own encode/decode logic; this centralizes them into jvm_types::modified_utf8 (operating on UTF-16 code units, no_std/alloc), reducing duplication and aligning behavior across the class loader and stream APIs.

Changes:

• Add jvm_types::modified_utf8 with encode(&[u16]) -> Vec<u8> and decode(&[u8]) -> Option<Vec<u16>>, preserving unpaired surrogates and rejecting malformed input; make jvm-types #![no_std] with extern crate alloc.
• Reuse the codec in classfile parsing (now intentionally rejecting standard 4‑byte UTF‑8) and in DataInputStream.readUTF / DataOutputStream.writeUTF, keeping the stream exceptions and 65,535-byte length limit.
• Expand tests for exact encoded bytes, unpaired surrogates, malformed sequences, and the encoded-length limit.

File	Description
jvm-types/​src/​modified_utf8.rs	New shared modified UTF-8 encode/decode over UTF-16 code units.
jvm-types/​src/​lib.rs	Makes crate #![no_std], adds alloc, exposes modified_utf8 module.
classfile/​src/​constant_pool.rs	Replaces inline decoder with shared codec; drops the 4-byte compatibility path; updates tests.
rustjava-runtime/​src/​classes/​java/​io/​data_input_stream.rs	readUTF now decodes via the shared codec.
rustjava-runtime/​src/​classes/​java/​io/​data_output_stream.rs	writeUTF now encodes via the shared codec.
rustjava-runtime/​tests/​classes/​java/​io/​test_data_output_stream.rs	Expands round-trip, malformed-input, and length-limit coverage.

I verified the encode/decode logic against the test vectors (all 12 code units, the 27-byte encoded length, NUL→C0 80, surrogate handling, and every malformed case), confirmed jvm-types has no std:: usage so the #![no_std] switch is safe, confirmed the removed imports (Vec, REPLACEMENT_CHARACTER) are no longer referenced, and confirmed no leftover callers of the old decode_modified_utf8. No objective issues were found.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.