[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 #2309, per the ruling (take the recommendation). Breaking: lifetime is renamed with no alias.

• Fix: RFC 8003 §4.5 lays out REG_FAILED as a Failure Type octet followed by Reg Type octets, with no lifetime. pcapkit read the first octet as a lifetime timedelta and decoded the Reg Types through the RegistrationFailure registry.
  • The schema, data model and maker now carry failure_type (a RegistrationFailure) in place of lifetime.
  • reg_type and reg_failed decode through the Registration registry.
  • The wire octets are unchanged.
• Tests: test_hip_roundtrip_unit pins the parsed types, parse→rebuild over Failure Types 0–255 with several Reg Types, and make→parse octets.
• Fail-without-fix: on origin/main, 7 failed, e.g. '03a8000300030100' != '03a8000302030100'. On the head, 44 passed.

[JarryShaw]

Review: GOOD TO GO at c75c8f120 (Sonnet).

• RFC 8003 §4.5, checked against the RFC text: REG_FAILED is a Failure Type octet followed by Reg Types, with no lifetime. failure_type uses RegistrationFailure, and reg_failed uses Registration, like REG_INFO, REG_REQUEST and REG_RESPONSE.
• Byte-exact: parse→rebuild over Failure Types 0, 1, 2, 3, 9, 201 and 255, plus make→parse. No lifetime attribute or keyword remains, and no leftover reference was found repo-wide.
• Changelog: a breaking-change bullet is in 1.5.0.rst and CHANGELOG.md. The PR body's changelog box now says so.
• Fail-without-fix: on origin/main, 7 failed across both test files (4 from the round-trip file alone). On the head, 51 passed.

[JarryShaw]

Rebased c75c8f120 → 4ec3e1d17 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, 4ec3e1d17, 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	17438	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 4ec3e1d17 → fa1dd3c83 onto main b7b02c0fa. The patch-id is identical to the old head, so the rebase applied every hunk — including the CHANGELOG.md/1.5.0.rst additions — with no textual conflict against #2312's entry.

The old head's CI was green (Required checks passed SUCCESS, 0 failures, KTLO clean, protocols ordering leg 1946 tests / 0 failures). On the new head: test_hip_roundtrip_unit.py, test_hip_unit.py, tests/test_docstring_contract.py and tests/test_tier_guard.py give 187 passed, and util/changelog_md.py --check passes.

The patch is unchanged and the old head was already green, so the owner can bypass checks and merge once the new head's CI completes.

[JarryShaw]

Correction to my own comment above: you had already merged this (as 9dd46d8d4) while I was rebasing it, so the "rebase" and the force-push after it landed on an orphan branch ref and changed nothing merged. Sorry for the noise — no action needed here.