refactor(protocols): one home per duplicated RFC-fixed constant (#1739)

[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

---

Description

Closes #1739, which carries ruling 7 from #1726. Every RFC- or format-fixed constant that was spelled out in more than one module now has one home. That home is the module that owns the protocol, or its schema module, and every other place reads the constant from there. None of them is configurable.

• Homes:

  • IPv4, IPv6, AH and TCP: the header lengths and offsets, plus IP_MAX_LENGTH;
  • PCAP: the header and record lengths, the magic number and the resolution;
  • PCAP-NG: the byte-order magic, the block minimums and the default resolution.

  Each home carries versionadded:: 1.5.0 and an autodata entry.

• No re-exports (refactor(corekit)!: move the byte-order field helpers into corekit (#1711) #1715, docs: #719-style prose sweep after the #1090–#1102 defect wave #1103): the old spellings are deleted. None of them shipped in 1.4.x, so they get no versionremoved.

• Guard, tests/project/test_rfc_constants_one_home_1739_unit.py: sweeps protocols, foundation, toolkit and dumpkit for every integer literal with a homed value, wherever it appears. Only the homes' own definitions are skipped. Every hit that is not a copy is allowlisted with a reason: 490 hits, 0 offenders.

• Edge test, tests/foundation/reassembly/test_ip_max_length_edge_1739_unit.py: pins IP_MAX_LENGTH as a strict bound.

Tests, narrow runs on Python 3.14:

• the five *_1739_unit.py files plus test_unit_tests_queue.py: 69 passed, 675 subtests;
• tests/foundation/reassembly: 134 passed.

Out of scope: the dpkt engine's loopback address-family keys repeat Loopback.__proto__. Those numbers are set by the operating systems, not by an RFC, so they are not covered by #1739.

[JarryShaw]

Verdict: GOOD TO GO at 9155375a6. A Sonnet cross-review (author: Opus) ran over four rounds.

• Round 4 blocker, now fixed: the guard's sweep depended on where a literal sat, so copies placed as a default argument, a tuple member, a conditional branch or a table entry went unflagged. It now flags every homed integer literal wherever it appears. The reviewer's own scan, written independently of the guard, matched the guard's 490 hits line for line.
• Mutants, all 41 killed: M1–M12, R1–R12, N1–N12 and X13–X17, plus 16 placements the reviewer added (negative and hex literals, keyword and lambda defaults, comprehensions, annotations and others). M11 is killed by the new IP_MAX_LENGTH edge test, not written off as equivalent.
• Allowlist: 12 reasons spot-checked against the code, and all of them match.
• Non-blocking notes:
  • the commit body runs about 22 lines;
  • a copy placed outside the four swept trees would not be caught, and none exists today.
• My own check: the five *_1739_unit.py files plus test_unit_tests_queue.py give 69 passed and 675 subtests, with pcapkit.__file__ printed inside the PR's worktree.

[github-actions]

Coverage: 89.59% (unit tier, Python 3.15, 9155375a6, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	3	20	3	95.16%
pcapkit/const	18860	954	2368	868	91.06%
pcapkit/corekit	2457	69	780	35	96.54%
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	17186	224	4574	189	98.08%
pcapkit/toolkit	872	177	298	8	77.18%
pcapkit/utilities	512	2	158	4	99.10%
pcapkit/vendor	4417	2342	1010	157	43.38%

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