[remislp]

dcio has never had CI. Nothing but a local test run has ever gated a change,
and it is now becoming the record layer under SCALCS, EKDIST and HJCFIT — a
regression here reaches three other projects.

The suite runs in under two seconds, so the matrix covers every supported
Python (3.11–3.13) and all three platforms rather than one representative
build. Reading and writing SCN means struct packing and endianness, which is
exactly the kind of thing that is fine on the machine it was written on.

Three gates beyond running pytest

Example records present. Several tests take a legacy SCN from examples/
and pytest.skip when it is absent. A skip is a silent pass: if those files
ever stopped being checked out, the golden burst counts and the round-trip
checks against real records would quietly stop running while CI stayed green.
The step asserts they are there rather than trusting a skip to be noticed.

Oldest dependencies. numpy>=2.0 is declared, but nothing has ever run
against 2.0 itself, so the floor is an assertion rather than a tested fact.
NEP 50 changed scalar promotion and impose_resolution does float arithmetic
on array elements, so a drifted minimum would show up there. That job pins
Python 3.11, because numpy 2.0 has no wheel for 3.13.

Wheel and console scripts. An editable install resolves imports and entry
points from the source tree, so it cannot catch an undeclared package or an
entry point that does not resolve. The three dcio-* commands have never been
run from an installed wheel.

Checked locally first

• All three console scripts answer --help with exit 0.
• Full suite passes on numpy 2.1 in a clean venv — the oldest available for
  this machine's Python 3.13. numpy 2.0 itself is what the oldest-deps job
  exists to find out about; it has not been run anywhere yet.

Sequencing

pull_request workflows run from the head branch, so #1 will not be checked by
this until it is rebased. Merge this first, then rebase bursts and #1 picks
up the full matrix.

🤖 Generated with Claude Code