fix(etag): avoid skipping headers when filtering 304 response headers - #5234

Hi @spellsaif,

Thanks for opening this PR. Nice catch! The proposed fix is functionally correct as written, and I don’t see any behavioral issue with it.

That said, how about using Array.from() to snapshot the header keys before deleting them? This avoids manually building headersToDelete and keeps the code shorter.

Both approaches allocate a temporary array. The performance difference may depend on the proportion of headers being deleted, but with realistic header counts, I expect it to be too small to measure in practice.

      for (const key of Array.from(c.res.headers.keys())) {
        if (retainedHeaders.indexOf(key.toLowerCase()) === -1) {
          c.res.headers.delete(key)
        }
      }

@usualoma Hi! Thanks for the review and suggestion. :)

I've updated the implementation to use Array.from(c.res.headers.keys()) as suggested. It keeps the code much cleaner and concise. All tests are passing green.

Thank you 👍

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 79.77%. Comparing base (41bdc42) to head (224a89c).

@@           Coverage Diff           @@
##             main    #5234   +/-   ##
=======================================
  Coverage   79.77%   79.77%           
=======================================
  Files         155      155           
  Lines       10934    10934           
  Branches     2292     2292           
=======================================
  Hits         8723     8723           
  Misses       2211     2211           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

@spellsaif @usualoma

Ah, this was actually a bug. Thanks!