[jasnell]

Filling more gaps

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 37.31%. Comparing base (8bfef8f) to head (02552e4).
⚠️ Report is 4 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #7182   +/-   ##
=======================================
  Coverage   37.31%   37.31%           
=======================================
  Files         800      800           
  Lines      251740   251740           
  Branches    20021    20021           
=======================================
  Hits        93943    93943           
  Misses     146447   146447           
  Partials    11350    11350           

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[guybedford]

Ran all readable-byte / writable / piping targets with --nocache_test_results: 55/55 pass. The piping additions and the WPT map look good (the 34 count is right). Approving, with a handful of comments where I probed both implementations and the comments/attributions don't match what the code actually does. None block, but the mislabelled spec claims are worth fixing since these files are the normative artifact for the suites.

Divergences I found while probing that are not yet pinned anywhere:

• Remainder after a partial BYOB read then default read: C++ copies into a fresh 4096-byte buffer (byteOffset 0), TypeScript returns a view into the original enqueued buffer (buffer.byteLength 16, byteOffset 8), which is exactly what the WPT asserts.
• byobRequest after a partial enqueue() into a pending pull-into: C++ invalidates it (null), TypeScript shrinks the view (byteLength 2 -> 1). enqueueDiscardsByobRequest only covers the full-fill case.