[JarryShaw]

• Searched for similar pull requests
• Followed the coding style
• make test passes, and a test case covers the change
• 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 — bug fix

---

Description

Closes #2362.

• Problem: _make_timestamp picked a defaulted timestamp's clock by reading self.nanosecond, which resolves its interface through the parsed data model. A block being made has no such model yet, so make() without an explicit timestamp raised AttributeError: '_info' for every block that carries one — the Enhanced Packet Block, the obsolete Packet Block, and the Interface Statistics Block, through its own Timestamp field as well as its isb_starttime/isb_endtime options.
• Fix: read the resolution from _get_resolution(interface_id) instead — the lookup the method already made for the scale; the offset keeps its own _get_offset(interface_id) — no new fallback, matching every other make-path lookup in this file.
• Test: tests/protocols/misc/test_pcapng_epb_pb_default_timestamp_2362_unit.py covers all three block kinds defaulting their timestamp to the current time, scaling to the interface's declared resolution and offset, reading the correct interface rather than interface 0 in a multi-interface section, reading the nanosecond clock rather than the float one when resolution calls for it, and round-tripping. Each case raises AttributeError on the parent commit.

Note: make() of EPB/PB/SPB with ctx=None still raises a separate, pre-existing UnsupportedCall on _get_linktype — a different symptom in a different method, out of scope here; worth its own issue if not already tracked.

[JarryShaw]

Review: GOOD TO GO at bcdc8f83e (Sonnet). Fix confirmed correct and minimal — 482 explicit-timestamp cases byte-identical before/after, 6 resolutions + offset confirmed via independent struct decode, round-trips exact, mutation-verified (13 failures restoring the bug). ctx=None caveat confirmed real and separate (pre-existing, in the constructor's re-parse step, not pack/make).

Two corrections to my own commit message/PR body: "ISB via isb_starttime/isb_endtime" understates it — the ISB block's own Timestamp field crashes too, not just its options. And "the same lookup already made two lines later for the offset" is wrong — the pre-existing lookup reused for the scale is _get_resolution; the offset uses the separate _get_offset. Will fix both.

Two mutation survivors found (interface_id hard-coded to 0, and the time_ns-vs-float clock branch unpinned) — fix dispatched, plus the docstring-tense nit. Issue label flipped pending→wip.

[github-actions]

Coverage: 91.58% (unit tier, Python 3.15, 384a4f646, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	2	95.16%
pcapkit/const	19601	881	2620	867	92.12%
pcapkit/corekit	2596	74	818	37	96.51%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	3029	23	1050	25	98.82%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17456	218	4598	185	98.15%
pcapkit/toolkit	914	36	334	10	95.51%
pcapkit/utilities	502	7	144	5	98.14%
pcapkit/vendor	4432	1988	1018	167	51.87%

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

[JarryShaw]

Amended bcdc8f83e → 83d72341a: added 2 mutation-verified tests (multi-interface resolution, nanosecond-vs-float clock pinning), reworded the docstring to the #1103 convention, and corrected my own commit message's two slips — ISB's own Timestamp field crashes too (not just its options), and the offset reuses a different lookup (_get_offset) than the scale does, not "the same lookup". Issue label flipped pending→wip. 8/8 tests pass.

All findings now addressed.

[JarryShaw]

Amended 83d72341a → 384a4f646: fixed the "Fix:" sentence (the offset keeps its own separate lookup, not "two lines later" — the real line gap is 12, not 2) and the remaining past-tense clause in the source docstring ("made a make... raise" → "would make... raise"). Prose-only, 8/8 tests pass.

All findings across both rounds now addressed.