Insert multiple values into a string pandas

Question

Suppose we have a string and a dataframe like this:

import pandas as pd

myStr="Insert {var1} here & here {var1} and insert {var2} here & here {var2}."
df = pd.DataFrame({'Var1': (1, 2), 'Var2': ('B', 'C')}

>>> df
   Var1 Var2
0     1    B
1     2    C

Now I add the myStr as a new column and I wanna insert the the var 1 and var2 their locations within the myStr. I am looking to get something like below. I was thinking to use f-strings but not sure what is the best way to do it in pandas.

Var1 Var2    new_col
1    B       Insert 1 here & here 1 and insert B here & here B.
2    C       Insert 2 here & here 2 and insert C here & here C.

score 2 · Answer 1 · answered Jun 15 '20 at 23:03

This is one way about it; create a list comprehension and generate a new dataframe - it should be easier and faster than using apply :

outcome = [(var1,var2,
            f"Insert {var1} here & here {var1} and insert {var2} here & here {var2}.")
            for (var1, var2) in df.to_numpy()]

pd.DataFrame(outcome,columns=["Var1","Var2","new_col"])

    Var1    Var2    new_col
0   1   B   Insert 1 here & here 1 and insert B here & her...
1   2   C   Insert 2 here & here 2 and insert C here & her...

Alternatively, using the myStr variable, in combination with the format method :

outcome = [(var1,var2, myStr.format(var1=var1,var2=var2))
            for (var1, var2) in df.to_numpy()]

Let's see the contents of outcome :

print(outcome)

[(1, 'B', 'Insert 1 here & here 1 and insert B here & here B.'),
 (2, 'C', 'Insert 2 here & here 2 and insert C here & here C.')]

Create dataframe, similar to the method above:

 pd.DataFrame(outcome,columns=["Var1","Var2","new_col"])

Augusto Sisa · Answer 2 · 2020-06-15T23:18:41.147

2

Use apply and axis parameter:

df['new_col'] = df.apply(lambda x: myStr.format(var1=x['Var1'], var2=x['Var2']), axis=1)

The axis parameter iterates over rows.

edited Jun 15 '20 at 23:18

answered Jun 15 '20 at 23:09

Augusto Sisa

548
4
15

score 1 · Answer 3 · answered Jun 16 '20 at 02:01

1

Another one-liner:

df['new_col']=[f'Insert {i[0]} here & here {i[0]} and insert {i[1]} here & here {i[1]}.' for i in df.values]

answered Jun 16 '20 at 02:01

Jack Fleeting

24,385
6
23
45

score 0 · Answer 4 · answered Jun 15 '20 at 23:14

0

Probably not the fastest way, but you didn't say you were working with anything big.

Axis=1 makes the apply go row by row.

df.apply((lambda row: row['new_col'].format(var1=row['Var1'], var2=row['Var2'])), axis=1)

answered Jun 15 '20 at 23:14

tomaszps

66
1
6

Insert multiple values into a string pandas

4 Answers4