[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 pylint not run; mypy is clean on the three UDP files; isort flags two TransportProtocol import lines that were already on main and that this PR leaves as is

• make test passes, and a test case covers the change — the whole suite was not run (see below); a new test module covers the change

• 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 of your pull request and other information

Closes #1461

UDP's payload now stops at its Length field, and the octets after it are info.trailer (present only when non-empty), as IPv4/IPv6 (#1431), IPX (#1435) and L2TPv2/OSPF (#1460) do. make(trailer=) writes them outside the declared length. !: on a datagram with excess octets, the payload no longer includes them.

Repro 0035003500080000 + pp: before, from_data(info) gave 8 of 10 octets; now 10/10, through info and through info.to_dict(). A Length below 8, of 0, or past the data still rebuilds exactly with no trailer. pcapkit does not compute the UDP checksum, so that path is unchanged. examples/captures: per-frame info is identical on all 1616 frames, and none gains a trailer.

Tests: tests/protocols/transport/test_udp_trailer_kept_unit.py fails 5 cases before the fix. On 94ae9664c + this commit: protocols --exclude protocols/internet leg 1156 tests, 0 failures, 0 errors; test_capture_roundtrip_runtime.py + tests/project + the new module 427 passed, 1 skipped.

[github-actions]

Coverage: 89.56% (unit tier, Python 3.14, 71b7a219c, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	4	93.55%
pcapkit/const	18795	957	2356	870	91.00%
pcapkit/corekit	2043	89	636	25	94.77%
pcapkit/dumpkit	242	2	88	2	98.79%
pcapkit/foundation	2648	98	930	48	94.80%
pcapkit/interface	112	7	40	5	92.11%
pcapkit/protocols	16898	222	4564	187	98.06%
pcapkit/toolkit	539	74	168	3	84.58%
pcapkit/utilities	429	4	122	4	98.55%
pcapkit/vendor	4409	2342	1006	157	43.25%

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

[JarryShaw]

Verdict: GOOD TO GO at 71b7a219c (Sonnet cross-review; the author ran on Opus).

• 60 hand-built datagrams rebuild byte-exact through info and to_dict(). They cover Length 0, 1, 7–12, 20 and 65535, payloads of 0, 1 and 8 octets, with and without a trailer. A trailer appears only when there are octets past Length.
• make(trailer=) writes outside the declared length.
• 16 Ethernet/IPv4/UDP frames are exact. They cover ports 53, 67, 1701 and 9999, each with and without a trailer on both the UDP and IPv4 layers. The upper layer gets exactly Length − 8 octets.
• I re-ran the issue repro myself: 10 of 10 octets, trailer=b'pp', in both forms.
• The captures are unchanged. Per-frame info hashes match main for all 23 captures (1616 frames, 103 UDP). Nothing in foundation/ or application/ reads the UDP payload.
• The tests fail without the fix: 5 failures with the UDP files taken from main.
• Legs: rest 1156, internet 617, and capture/project 427, all passing. The transport edge module passes.
• The ! is justified, and nothing builds Data_UDP directly.