[JarryShaw]

• 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

• Added a changelog entry — N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

• docs — documentation only

Per the ruling on #1921, this drops :meta private: from the get overrides of Command, Method, OptionType, TransportProtocol and AppType. Every :meth:get`` reference on the ftp, http, pcapng and reg pages now links.

The const modules are regenerated from the edited templates, reusing the crawled parts already committed, and are byte-identical to the rendered output. Each summary now states how the lookup handles case and what a miss returns. OptionType.get also gets a full description of its lookup rules.

A new project test fails if an override, its template, or a :meth: reference to one is marked private. The pinned docs build reports 0 warnings.

#1910 changed some AppType.get references to plain literals. They can link again now, but this PR leaves them as they are.

Closes #1921

[JarryShaw]

NEEDS CHANGES at 9b6b2514b. A Sonnet cross-review called every override, and the case handling, lookup order and unchanged table sizes on a miss all match the docstrings. Every get() chip on the four pages links. Two small fixes are still needed. First, AppType.get says a miss is "named after the IANA range holding the port". In fact it takes the label of the span holding the port (reserved, unassigned, or a service name), and unknown in a gap, so say exactly that. Second, the new test's prose carries timed context: "no longer overrides get" in an assertion message, and "carried" in the module docstring. Restate both as what the test checks.

[JarryShaw]

GOOD TO GO at b20a6a62c. The miss wording now matches a sweep of every TCP port. Unregistered ports get reserved, unassigned, a service label (ircu at 6666, swx at 7300), or unknown. Port 300 is a registered port (tacacss), not a miss, so the earlier review's example was wrong. The template and const hunks are byte-identical. The test prose has no timed context. Dropping a pinned :meth:get`` now fails the test.