Actionable comments posted: 0

🧹 Nitpick comments (1)

crates/stdlib/src/ssl/compat.rs (1)

424-466: Core fix looks correct, but consider documenting the buffer size choice.

This retry logic properly handles the case where rustls reports it doesn't need more data but additional TLS records may still arrive from the network. The distinction between non-blocking (immediate return) and blocking/timeout sockets (attempt read) is correct.

Minor observations:

• Line 439: The hardcoded 2048 buffer size is reasonable for TLS records but lacks explanation. Consider adding a comment explaining why this size was chosen (e.g., typical TLS record overhead + payload).
• Line 449-453: The clone() and unwrap_or(0) are safe but slightly inefficient. The clone is necessary for try_into_value, and the unwrap_or provides a safe fallback if conversion fails.

💡 Optional: Add explanatory comment for buffer size

-    let data = match socket.sock_recv(2048, vm) {
+    // Read up to 2048 bytes (sufficient for typical TLS record: 5-byte header + up to ~16KB payload fragment)
+    let data = match socket.sock_recv(2048, vm) {

📜 Review details

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between ebdc033 and 709ba76.

📒 Files selected for processing (1)

• crates/stdlib/src/ssl/compat.rs

🧰 Additional context used 📓 Path-based instructions (1) **/*.rs

📄 CodeRabbit inference engine (.github/copilot-instructions.md)

**/*.rs: Follow the default rustfmt code style by running cargo fmt to format Rust code
Always run clippy to lint Rust code (cargo clippy) before completing tasks and fix any warnings or lints introduced by changes
Follow Rust best practices for error handling and memory management
Use the macro system (pyclass, pymodule, pyfunction, etc.) when implementing Python functionality in Rust

Files:

• crates/stdlib/src/ssl/compat.rs

🧬 Code graph analysis (1) crates/stdlib/src/ssl/compat.rs (2)

crates/stdlib/src/ssl.rs (2)

• timeout (4165-4168)
• e (1115-1115)

crates/stdlib/src/socket.rs (4)

• timeout (707-709)
• std (1538-1538)
• std (1598-1598)
• new (2446-2450)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Run snippets and cpython tests (windows-latest)

🔇 Additional comments (4)

crates/stdlib/src/ssl/compat.rs (4)

399-399: LGTM: Timeout variant appropriately distinguishes timeout errors.

The new Timeout(String) variant properly separates socket timeout scenarios from generic IO errors and WantRead conditions, improving error clarity and matching Python's timeout exception semantics.

---

479-481: LGTM: Simplified fallback defers to main buffer handling.

The simplified continue is appropriate here since the primary buffer-full logic is now handled in ssl_read_tls_records (lines 569-585). This serves as a safety net fallback.

---

560-599: Excellent buffer-full handling with proper safeguards.

The enhanced retry logic correctly handles rustls buffer exhaustion:

1. When read_tls returns 0, calls process_new_packets to consume buffered data and make room
2. Retries once to feed remaining data
3. Critical: Line 576-577 breaks if still can't consume, preventing infinite loops

The dual handling of Ok(0) and Err("buffer full") paths ensures robustness across different rustls error modes.

---

656-659: LGTM: Correct semantic distinction between timeout and would-block.

Changing from WantRead to Timeout when the socket times out during select is semantically correct. WantRead indicates a non-blocking socket would block, while Timeout indicates an operation exceeded its time limit. This properly aligns with Python's socket.timeout exception behavior.