Skip to content

pandas/empty-guard

Rule empty-guard
Pack pandas
Severity warning
Fix unsafe, with xpd check --fix --unsafe-fixes
Library versions every version

The mean, median and similar reductions of an empty Series are already NaN, with no error. Drop the test for an empty frame.

share = (capped["upper"] < LOW_PRICE).mean() if len(capped) else float("nan")
share = (capped["upper"] < LOW_PRICE).mean()

The rule reports x if len(frame) else nan where x is mean, median, std, var, min, max or quantile of a column picked with a string key. It leaves sum alone, because the sum of an empty Series is 0, and it leaves a NumPy array alone, because np.mean of an empty array warns.

xpd check --fix --unsafe-fixes applies the fixes of this rule. Check the result, because an unsafe fix can change what the code does.

The unsafe fix writes the reduction without the test, because the frame in the test and the frame in the reduction can differ.

xpd applies the replacement with --fix --unsafe-fixes.

With warnings turned into errors, mean, median, std, var, min, max and quantile(0.5) of pd.Series([], dtype=float) each returned nan on pandas 3.1.1. The test repeats what pandas does, and a reader has to check that both branches give the same answer.

xpd prints this message with the issue. A name that starts with $ stands for code from your file.

$VALUE is nan for an empty Series already, with no error and no warning. Drop the test for an empty frame.

Browse the other rules of the pandas pack.