[JarryShaw]

Please follow the guide below

• You will be asked some questions, please read them carefully and answer honestly

• Put an x into all the boxes [ ] relevant to your pull request (like that [x])

• Use Preview tab to see how your pull request will actually look like

• Searched for similar pull requests

• Followed the coding style (make pylint, make mypy, make isort) — mypy on ipv6_ext.py, isort and pylint E/W on the new test only; no new messages, but not the full make targets

• make test passes, and a test case covers the change — per module only, listed below; not the full make test

• Added a changelog entry under docs/source/changelog/ and regenerated CHANGELOG.md, if the change is user-visible — N/A, added centrally

What is the purpose of your pull request?

• fix — corrects a defect

---

Description of your pull request and other information

Closes #1178 (option 2). IPv6_Ext.__length_hint__ is now 8: RFC 6564 §4's Hdr Ext Len excludes the first 8 octets, and RFC 8200 §4 makes headers multiples of 8. All 8 subclasses (AH 20, HIP 40, ESP/HOPOPT/IPv6_Frag/IPv6_Opts/IPv6_Route/MH 8) define their own hint, so none inherits it.

_parse_next_layer(IPv6_Ext, tail, 300), alias absent/200/Shim6/HOPOPT	before	after
2–7 octets	IPv6_Ext, length 8	Raw, rebuilds the tail exactly
8, 16 octets	IPv6_Ext	unchanged

The IPv6 chain walk hands each header its captured slice, so it never met this guard; its info for HOPOPT/Shim6/Opts/Route tails of 2–16 octets is identical before and after.

Tests: new test_ipv6_ext_hint_unit.py 2 passed / 32 subtests; with the hint reverted to 2, 24 subtests fail. The pin is renamed test_length_hint_is_the_eight_octet_rfc_6564_minimum. ipv6_ext 30, ipv6_extension 61, length_hint_minimum_header 4, next_layer_short_header 7, option_roundtrip 6, tests/project 389 passed / 1 skipped. Ordering leg protocols/internet: 474 tests, 0F/0E.

[JarryShaw]

Cross-review verdict on 2296f848a: GOOD TO GO (ran on Sonnet; author Opus). Labelled review: good-to-go.

• RFC basis: RFC 6564 §4 Figure 1 and RFC 8200 §4 both put the minimum at 8 octets. The reviewer checked both against the RFC text.
• No subclass inherits the base hint: all 8 subclasses define their own (__subclasses__() walk), with HIP at 40, AH at 20 and the other six at 8.
• No regression: the capture sweep over examples/captures is identical against main (1602 packets, 1422 round-trip), and so is a targeted probe of tails of 1–24 octets for protocols 0/60/140/253/254/43/135/139/252, over IPv4 and IPv6 alike.
• Scope of the effect: none is observable through IPv6(...), since the chain walk sizes each header to the octets it captured. The fix bites only when a caller over-declares the length to _parse_next_layer, which is what the PR body's table already says.
• Tests: the new test fails without the fix (24 subtests) and imports per test. The protocols/internet ordering leg ran 474 tests with 0 failures and 0 errors.
• Unverified: pcapkit.extract on a crafted truncated capture. It shares the IPv6 walk, so it should behave the same, but that is inferred rather than run.

[github-actions]

Coverage: 88.89% (unit tier, Python 3.14, 2296f848a, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	4	93.55%
pcapkit/const	18757	1034	2342	845	90.24%
pcapkit/corekit	1878	91	580	22	94.34%
pcapkit/dumpkit	142	0	42	0	100.00%
pcapkit/foundation	2442	102	852	39	94.38%
pcapkit/interface	112	7	40	5	92.11%
pcapkit/protocols	15707	188	3974	163	98.19%
pcapkit/toolkit	487	65	144	1	85.42%
pcapkit/utilities	429	4	122	4	98.55%
pcapkit/vendor	4409	2359	1006	158	42.84%

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