Is your feature request related to a problem? Please describe.
Fugue always requires dataframe-like input and output for transformers. But sometimes it is inconvenient because sometimes the best semantic of transformation would be a row to a dataframe or a dataframe to a row or a row to a row. So supporting single row functions can further reduce the frictions to express your logic in the best way.
def my_funct(row:Dict[str,Any]) -> dict[str,Any]:
row["b"]=2
return row
import fugue.api as fa
import pandas as pd
df = pd.DataFrame(dict(a=[1,2,3]))
fa.transform(df, my_func, schema="*,b:long")
Is your feature request related to a problem? Please describe. Fugue always requires dataframe-like input and output for transformers. But sometimes it is inconvenient because sometimes the best semantic of transformation would be a row to a dataframe or a dataframe to a row or a row to a row. So supporting single row functions can further reduce the frictions to express your logic in the best way.