[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

What is the purpose of your pull request?

• fix — corrects a defect

---

Description of your pull request and other information

Closes #2132.

• Bug: when the pcap module is installed but cannot load (OSError from its loader), extract(engine='pypcap') raised the error. engine='pcap_ct' warned and fell back. PyPCAP._reason_for only checked wrong_backend_reason.
• Fix: both engines' _reason_for now call a new _pcap_backend.engine_reason(wanted, found), so they cannot drift apart. It returns the wrong-backend reason, or PCAP_CT's existing "installed but unusable -- " wording, or None when the module is absent, which the import test already reports.
• Constructor: the PyPCAP.__init__ backstop raises UnsupportedCall rather than letting the OSError out of import pcap.
• Docs: an engine_reason entry in 3rdparty.rst, and the "declines to run" bullet in engines/index.rst.
• Tests: tests/foundation/engines/test_pypcap_unusable_module_2132_unit.py.
  • It covers the helper, plus a child interpreter with a fake pcap that raises OSError('Cannot find libpcap.so library'). In the child, both engines must warn and fall back.
  • Against main: 8 failed, 2 passed. The passing pcap_ct case is the control.
  • With the fix: 8 passed. The foundation ordering leg gives 738 tests, 0 failures.

[JarryShaw]

GOOD TO GO at fe72cfa03 (Sonnet review of cdff40b57, plus its one nit, now applied).

• Shared verdict: _pcap_backend.engine_reason keeps PCAP_CT's verdicts exactly; I diffed the old body against the new. wrong_backend_reason still takes precedence. An absent module still returns None for the import test.
• Extraction: Extractor warns and falls back on the new reason before PyPCAP.__init__ runs. The __init__ backstop's UnsupportedCall is visible only to someone constructing the engine directly.
• Tests: 6 of the 8 new tests fail on main's three engine files. The pcap_ct end-to-end case passes there, as the control. The child-interpreter test asserts it runs the PR's pcapkit.
• Nit applied: fe72cfa03 changes only engine_reason's docstring. It drops the history sentences and keeps the contract with :issue:`2132` . The range-diff shows nothing else.

[github-actions]

Coverage: 91.55% (unit tier, Python 3.15, fe72cfa03, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	2	95.16%
pcapkit/const	19599	880	2618	866	92.12%
pcapkit/corekit	2589	75	816	37	96.48%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	2971	23	1036	25	98.80%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17418	219	4574	185	98.14%
pcapkit/toolkit	869	36	298	10	95.20%
pcapkit/utilities	498	7	144	5	98.13%
pcapkit/vendor	4432	1988	1018	167	51.87%

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

[JarryShaw]

Ready to merge at fe72cfa03.

• Review: the Sonnet review is GOOD TO GO, and its docstring nit is applied.
• CI: every blocking check passed (Required checks passed SUCCESS), and the 3.8–3.10 KTLO legs carry no warnings.
• Ordering: the head sits directly on main 8fdc2fb33. Its tests/foundation plain-unittest ordering leg ran 738 tests with 0 failures, on the code this head carries; only a docstring changed since.

This closes #2132.