How do I remove text from strings in all the rows in a column?-CodePudding

Problem

I want to remove some surrounding text from strings in a dataframe.

Reprex

What I have

df1 = pd.DataFrame({'a': [1, 1, 2, 2, 3],
                    'b': ["NOSE PARKER Bond 1 Spain", "Fire PA1KER Bond 10 UK", 
                          "NOSE 2HANDS Bond 3 FRANCE", "EARS STARKER Bond 11 SOUTH AFRICA",
                          "NORSEPACKER Bond 01 JAPAN2002"],
                    'c': [13, 9, 12, 5, 5]})
df1

What I want

df2 = pd.DataFrame({'a': [1, 1, 2, 2, 3],
                    'b': ["NOSE PARKER Bond 1 Spain", "Fire PA1KER Bond 10 UK", 
                          "NOSE 2HANDS Bond 3 FRANCE", "EARS STARKER Bond 11 SOUTH AFRICA",  
                          "NORSEPACKER Bond 01 JAPAN2002"],
                    'c': [13, 9, 12, 5, 5],
                    'b new': ["Bond 1", "Bond 10", "Bond 3", "Bond 11", "Bond 01"]})
df2

My Attempt

df1["b new"] = df1["b"].str.replace("[^Bond\s\d]", "").str.strip()
df1

CodePudding user response：

you can use extract method to get required format.As per below regex there will be a space before bond and space before digit and digit can be one or more.

df['b_new'] = df.b.str.extract('( Bond \d )')

Result:

   a                                  b   c     b_new
0  1           NOSE PARKER Bond 1 Spain  13    Bond 1
1  1             Fire PA1KER Bond 10 UK   9   Bond 10
2  2          NOSE 2HANDS Bond 3 FRANCE  12    Bond 3
3  2  EARS STARKER Bond 11 SOUTH AFRICA   5   Bond 11
4  3      NORSEPACKER Bond 01 JAPAN2002   5   Bond 01