[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)

• make test passes, and a test case covers the change. Narrow runs only, listed below.

• Added a changelog entry under docs/source/changelog/ and regenerated CHANGELOG.md, if the change is user-visible: N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

What is the purpose of your pull request?

Tick the commit type your subject line carries.

• fix — corrects a defect
• feat — adds a feature
• perf — changes performance, not behaviour
• refactor — changes neither behaviour nor performance
• test — tests only
• docs — documentation only
• ci — workflows or build tooling
• release — bumps the version or rolls up a distribution
• chore — anything else

---

Counts:

• Drop counted numbers from the NGAP, HTTP, HTTP/2, PCAP, PCAP-NG, schema
  base and protocol base docstrings and pages; a spec total used as a
  coverage claim becomes count-free, and measurements stay.

Rendered output:

• The OSPF and DSB selectors' Returns is one paragraph, not a list behind an
  empty paragraph, and the DSB one states what the selector returns.
• NGAP's conversion table keeps its last row whole, and the class entry sits
  above the narrative sections.
• Document the root data model and PCAPNG.packet, reattach doc comments a
  NOTE had cut off, and keep the stdlib enum hook off the enum pages.
• Fix dead or under-qualified cross-references, split napoleon types in the
  Protocol and Raw constructors, and doc-comment types that rendered raw
  aliases.

Accuracy and wording:

• RARP and ARP share a frame format, not an EtherType; readers return parsed
  data; index raises as documented; grammar and spelling fixes; no
  timed wording in the TLSKeyLabel removal note.

Directives:

• versionchanged on OSPF and HTTP/2 make() for a parameter inserted before
  existing ones; the OSPF and RARP removal notes name the package-level
  names that went.

Non-prose check: ast.dump is unchanged on every .py file once docstrings are blanked. The OSPF and HTTP/2 make() directives are checked against v1.4.1: both signatures move an existing positional parameter. Narrow test runs are green.

Refs #1103

[JarryShaw]

Verdict: NEEDS CHANGES at db9458d6b. A Sonnet cross-review (author: Opus) confirmed:

• the OSPF and HTTP/2 make() directives, by AST against v1.4.1;
• the OSPF/RARP removal names, by importing them at the head;
• that ARP (0x0806) and RARP (0x8035) differ in EtherType;
• every "Parsed" return, the __index__ Raises, the NGAP table (docutils), the DSB selector's runtime behaviour, and the cross-reference targets.

Two blocking defects, both re-checked:

• protocol.py:1247-1248 says packet is an out-of-band keyword "consumed here". __init__ pops _layer, _protocol and __context__, but only reads packet (kwargs.get('packet')), which is forwarded.
• The :exclude-members: +_generate_next_value_ lines in httpv2.rst and pcapng.rst have no effect. The hook still renders in a clean build of this branch, because napoleon's skip handler overrides exclude-members.

A worker is adding a fix commit to this branch.

[JarryShaw]

Verdict: GOOD TO GO at e26048bcf. A Sonnet re-review of the fix commit confirmed:

• the ProtocolBase.__init__ note: _layer, _protocol and __context__ are popped, and packet is only read;
• the hook mechanism in Sphinx 9.1.0: emit_firstresult takes the first non-None result in ascending priority, autodoc applies exclude-members before the event, and napoleon includes the hook. Across every member of every pcapkit class (about 223k decisions), _generate_next_value_ is the only decision that changes, and it changes only at priority 400;
• that all 21 hook entries in the previous build now skip, while _missing_ stays documented on every enum;
• that the new test passes (7), fails when the connect line is removed or the priority is raised to 500, and its tree was restored clean;
• the OSPF, NGAP and "Drops count" polish against v1.4.1 and the code.

[github-actions]

Coverage: 90.05% (unit tier, Python 3.15, e26048bcf, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	2	20	2	96.77%
pcapkit/const	19595	881	2618	867	92.11%
pcapkit/corekit	2462	69	780	35	96.55%
pcapkit/dumpkit	276	3	100	3	98.40%
pcapkit/foundation	2915	164	1016	34	93.59%
pcapkit/interface	138	7	52	5	93.68%
pcapkit/protocols	17408	224	4580	189	98.10%
pcapkit/toolkit	872	177	298	8	77.18%
pcapkit/utilities	512	2	158	4	99.10%
pcapkit/vendor	4407	2342	1010	157	43.27%

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