[Code scan] ABACUS STRU scalar sc=False writes malformed position lines

[njzjz]

This issue is part of a Codex global repository code scan.

The ABACUS STRU writer has an operator-precedence bug when writing scalar spin-constraint (sc) flags. For False, the expression appends only "0" instead of appending " sc 0", so the 0 is concatenated directly to the preceding move flag.

Affected code:

dpdata/dpdata/formats/abacus/stru.py

Lines 797 to 805 in a7a50bf

	if sc is not None:
	if isinstance(ndarray2list(sc[iatomtype]), (list, tuple)) and len(
	sc[iatomtype]
	) in [1, 3]:
	iout += " sc " + " ".join(
	["1" if ii else "0" for ii in sc[iatomtype]]
	)
	elif isinstance(ndarray2list(sc[iatomtype]), (int, float, bool)):
	iout += " sc " + "1" if sc[iatomtype] else "0"

Minimal reproducer:

import dpdata

s = dpdata.System("tests/abacus.scf/STRU.ch4", fmt="stru")
s.to(
    "stru",
    "STRU_tmp",
    mass=[12, 1],
    pp_file=["C.upf", "H.upf"],
    sc=[False] * s.get_natoms(),
)
print(open("STRU_tmp").read())

Current position lines include malformed endings like:

5.192682633809 4.557725978258 4.436846615358 1 1 10
5.416431453540 4.011298860305 3.511161492417 0 0 00

Expected output should include an explicit sc 0 token rather than concatenating 0 to the previous field.

[dosubot]

Good catch. The root cause is a Python operator precedence issue on line 805:

iout += " sc " + "1" if sc[iatomtype] else "0"

Python parses this as:

iout += (" sc " + "1") if sc[iatomtype] else "0"

So when sc[iatomtype] is False, only the bare "0" is appended (no " sc " prefix), which concatenates directly onto the preceding move flag.

The fix is to parenthesize the ternary expression:

iout += " sc " + ("1" if sc[iatomtype] else "0")

No existing PRs or issues address this specific bug. A related fix was merged in PR #852 [1], which corrected per-atom indexing in the same function, but didn't touch this precedence issue.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.