[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:

• reword the SCTP PPID and error-cause notes, the register_apptype note,
  the HTTP port comments, the L2TP dispatch prose, the MPTCP subtype
  comment and the nested_length docstring without counting

Accuracy:

• L2TP no longer says only one version exists: L2TPv3 and L2F exist,
  unimplemented
• TCP Flags.urg is the flag, not the pointer; the option-list and
  unassigned-option docstrings say what they do; selectors document
  their FieldError
• the L2TPv2 note points at the module page, not a missing measurement

Directives:

• versionchanged 1.5.0 on TCP.make and UDP.make for out-of-range ports
  (const(apptype): AppType.get() swallows the range guard and mints out-of-range ports #758), and on UDP.make for total_length preceding payload (design(protocols): should from_data keep declared lengths on truncated captures? #1155)

Rendering:

• single-paragraph Returns on the MPTCP selectors, so no empty paragraph
• autodoc no longer reads #: comments' leading phrases as types
• drop the duplicate Flags stub on tcp.rst and repeated notes on
  l2tpv2.rst and s_tag.rst; move the L2TP class above its narrative
• document the ARP schema fields and l2tp_payload_length

Grammar: e.g./etc., register Warns and notes, reserved rows, init.

Non-prose check: ast.dump is unchanged once docstrings are blanked; any workflow, TOML or script edits are comment-only (parsed and compared). Narrow test runs are green (tests/project and the tests that read edited files).

Refs #1103

[JarryShaw]

Verdict: GOOD TO GO at acc1fda0d. A Sonnet cross-review (author: Opus) checked:

• the versionchanged:: 1.5.0 directives on TCP.make and UDP.make. 1.4.1's make() accepts port 99999 and the head raises ValueError, with 65535 still accepted. UDP's fourth positional argument moved from payload to total_length, as the directive warns;
• the L2TP wording against RFC 3931 and RFC 2341, and that Flags.urg is the flag bit;
• that the rewritten #: comments say the same thing;
• the removed duplicate Flags stub and notes, which lose no documentation;
• the new ARP field and l2tp_payload_length descriptions against RFC 826 and the code;
• both selectors' FieldError conditions, by running them;
• the count rewordings ("every error cause RFC 9260 §3.3.10 defines" covers causes 1 to 13). No count was added, and the ASTs are identical with docstrings blanked.

A combined clean Sphinx build of main with this branch adds no warnings. Non-blocking: in 1.4.1, the TCP(srcport=N) constructor raised AttributeError for any bare int, so the directive is accurate for make() only.