PERF: join empty frame by lukemanley · Pull Request #46015 · pandas-dev/pandas (original) (raw)

DataFrame.join already had a fast path for joining with an empty frame but it only covered a limited set of cases. This PR makes the fast path a bit faster and covers additional cases.

import pandas as pd 
import numpy as np

N = 10_000_000

df = pd.DataFrame({'A': np.arange(N)})
df_empty = pd.DataFrame(columns=['B', 'C'], dtype='int64')

%timeit df.join(df_empty, how='inner')
932 ms ± 83.4 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)       <- main
285 µs ± 4.29 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)  <- PR