Join two dataframes along columns with non-unique index

Question

I have two data frames that I want to join them along columns. The index is not unique:

df1 = pd.DataFrame({'A': ['0', '1', '2', '2'],'B': ['B0', 'B1', 'B2', 'B3'],'C': ['C0', 'C1', 'C2', 'C3']}):
    A   B   C
0  0  B0  C0
1  1  B1  C1
2  2  B2  C2
3  2  B3  C3

df2 = pd.DataFrame({'A': ['0', '2', '3'],'E': ['E0', 'E1', 'E2']},index=[0, 2, 3])
    A   E
0  0  E0
1  2  E1
2  3  E2

A should be my index. what I want is:

    A   B   C   E
0  0  B0  C0    E0
1  1  B1  C1    NAN
2  2  B2  C2    E1
3  2  B3  C3    E1

This pd.concat([df1, df2], 1) gives me error:

Reindexing only valid with uniquely valued Index objects

Answer 1

Maybe you're looking for a left outer merge .

df1.merge(df2, how='left')
   A   B   C    E
0  0  B0  C0   E0
1  1  B1  C1  NaN
2  2  B2  C2   E1
3  2  B3  C3   E1

Answer 2

By using combine_first

df1.combine_first(df2).dropna(subset=['A'],axis=0)
Out[320]: 
    A   B   C    D    E
0  A0  B0  C0   D0   E0
1  A1  B1  C1  NaN  NaN
2  A2  B2  C2   D1   E1
2  A3  B3  C3   D1   E1

After you edit:

By using combine_first

df1.combine_first(df2.set_index('A'))
Out[338]: 
   A   B   C    E
0  0  B0  C0   E0
1  1  B1  C1  NaN
2  2  B2  C2   E1
3  2  B3  C3   E2

Or

pd.concat([df1,df2.set_index('A')],axis=1)
Out[339]: 
   A   B   C    E
0  0  B0  C0   E0
1  1  B1  C1  NaN
2  2  B2  C2   E1
3  2  B3  C3   E2

Answer 3

concatenate along the column axis with concat

import pandas as pd

df1 = pd.DataFrame({'A': ['A0', 'A1', 'A2', 'A3'],'B': ['B0', 'B1', 'B2', 'B3'],'C': ['C0', 'C1', 'C2', 'C3']},index=[0, 1, 2, 2])

df2 = pd.DataFrame({'D': ['D0', 'D1'],'E': ['E0', 'E1']},index=[0, 2])

df = pd.concat([df1, df2], axis=1)

output:

    A   B   C    D    E
0  A0  B0  C0   D0   E0
1  A1  B1  C1  NaN  NaN
2  A2  B2  C2   D1   E1
2  A3  B3  C3   D1   E1