[JarryShaw]

• Searched for similar pull requests
• Followed the coding style
• make test passes, and a test case covers the change
• Changelog entry: 1.5.0.rst (breaking change)

What is the purpose of your pull request?

• fix — bug fix

---

Description

Closes #2311, per the ruling (take the recommendation). Breaking: the default error code changes.

• Fix: HTTP._make_http_rst_stream and HTTP._make_http_goaway defaulted error to HTTP_1_1_REQUIRED (0xd). A frame built without an error code therefore asked the peer to fall back to HTTP/1.1. Both now default to NO_ERROR (0x0), per RFC 9113 §7. The docstrings carry a versionchanged note.
• Changelog: 1.5.0.rst records the breaking change, and CHANGELOG.md is regenerated with util/changelog_md.py.
• Test: test_httpv2_error_default_2311_unit.py builds both frames with frame={}, pins the error-code octets as zero, and checks that an explicit error is still honoured.
• Fail-without-fix: on origin/main, 2 failed (the last octet is \r, not \x00). On the head, 3 passed.

[JarryShaw]

Review: GOOD TO GO at a90b72a43 (Sonnet).

• RFC 9113 §7, checked against the RFC text: NO_ERROR means the condition is not an error. That makes it the neutral default for both frames.
• Defaults: RST_STREAM and GOAWAY now default to NO_ERROR, and an explicit HTTP_1_1_REQUIRED still packs 0000000d. No other HTTP/2 maker keeps the old default.
• Changelog: the entry closes the "Changed" block in the file's breaking-change form, and util/changelog_md.py --check passes.
• Fail-without-fix: on origin/main, 2 failed (trailing \r). On the head, 144 passed, and the neighbouring HTTP/2 tests gave 118 passed.
• Non-blocking: the default frame=b'' still cannot build these frames. That is fix(httpv2): make's default frame=b'' cannot build frame types with a fixed-size payload #2274, which is blocked until this PR merges.

[JarryShaw]

Rebased a90b72a43 → 1f032f8ff onto main bfe184bd6. The patch-id is identical, so the GOOD TO GO verdict carries over to the new head.

The old head's Project tests failed on test_no_test_calls_unparse_outside_a_version_gate. That was not this PR. It came from main's #2306 scan test, which called ast.unparse outside a version gate; bfe184bd6 fixes it on main, and CI is re-running on the new head.

[github-actions]

Coverage: 91.56% (unit tier, Python 3.15, 1f032f8ff, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	2	95.16%
pcapkit/const	19599	881	2618	867	92.11%
pcapkit/corekit	2589	74	816	37	96.51%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	3023	23	1044	25	98.82%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17437	218	4586	185	98.15%
pcapkit/toolkit	869	36	298	10	95.20%
pcapkit/utilities	504	7	146	5	98.15%
pcapkit/vendor	4432	1988	1018	167	51.87%

Per-file detail: the coverage-html artifact of this run.

[JarryShaw]

Rebased 1f032f8ff → eb18a5907 onto main b7b02c0fa. The patch-id is identical to the old head — git's rebase applied every hunk, including the CHANGELOG.md/1.5.0.rst additions, with no textual conflict against #2312's changelog entry, so nothing in the diff changed.

The old head's CI was green (Required checks passed SUCCESS, 0 failures, KTLO clean). On the new head: tests/protocols/application (httpv2 and http), tests/test_docstring_contract.py and tests/test_tier_guard.py give 211 passed, and util/changelog_md.py --check passes.

Since the patch is unchanged and the old head was already green, the owner can bypass checks and merge once the new head's CI completes — this only re-establishes a path through the required contexts on top of main.

[JarryShaw]

Rebased again eb18a5907 → 325659a6a onto main 9dd46d8d4. This time it was a real conflict: #2309 (HIP REG_FAILED) landed on main between my last rebase and now, and both it and this PR append a "breaking change" bullet to the same spot in CHANGELOG.md and 1.5.0.rst. Resolved by keeping both bullets, in the order they already appear in 1.5.0.rst, then regenerating CHANGELOG.md from it with util/changelog_md.py rather than hand-editing the generated copy. git patch-id on pcapkit/protocols/application/httpv2.py and the test file is identical before and after — only the changelog resolution changed.

On the new head: tests/protocols/application (httpv2 and http), tests/test_docstring_contract.py and tests/test_tier_guard.py give 211 passed, and util/changelog_md.py --check passes.

My earlier note on this PR described a clean rebase against an older main; that rebase is superseded by this one. Not safe to bypass checks and merge until the new head's CI completes — flagging that explicitly since I said otherwise before #2309 landed.