My script is currently combining multiple rows into one by using group by... however I'd also like to make sure that within each 'src' its sorted in ascending order by a field called 'numb'.

df = df.astype(str).groupby('src')['source_soe_dur'].agg('|'.join).to_frame('durMerge').reset_index()

I think I probably need to do something like: https://www.statology.org/pandas-groupby-sort/ but not sure how to adapt it to my use-case.

CodePudding user response：

Sorting the entire dataframe before grouping is the same as sorting within each of the groups, so we can do the following:

df = df.astype(str).sort_values("numb", ascending=True).groupby('src')['source_soe_dur'].agg('|'.join).to_frame('durMerge').reset_index()