Name             Code
Ramesh           1234
Ramesh kumar       
Suresh           4532
Sur

This is the sample data ,if name contains similar string it should populate the code

Expected result:

Name             Code
Ramesh           1234
Ramesh kumar     1234 
Suresh           4532
Sur              4532

CodePudding user response：

Your question does not match your example and your two examples do not match. Here is an answer assuming that your first example is actually what you want - that you want to match the first three letters of the names to the code. Matching names can be a lot more complicated if you have many similar names, so this is not a particularly general solution, but it works for your example.

import numpy as np
import pandas as pd

#   Create test dataframe
test = pd.DataFrame({'Name':['Ramesh','Ramesh kumar','Suresh','Sur'],
                     'Code':[1234,np.nan,4532,np.nan]})

#   Create dictionary that maps first three letters of Name to value in Code
dfmap = test[~np.isnan(test['Code'])]
dictmap = {x['Name'][:3]:x['Code'] for i,x in dfmap.iterrows()}

#   Map the first three letters of the name to a new column
test['mapping'] = test['Name'].str[:3].map(dictmap)

#   Write over nans in the Code column with the values in the new column
test.loc[np.isnan(test['Code']),'Code'] = test.loc[np.isnan(test['Code']),'mapping']

#   Remove temporary mapping column
test = test.drop(columns='mapping')

CodePudding user response：

Considering your example and without further explanation, the .ffill() method could be enough:

df = (pd.DataFrame({'name' : ['Ramesh', 'Ramesh Kumar', 'Ram', 'Sur', 'Suresh Kumar', 'Suresh'],
                   'code' : [123,np.NaN,np.NaN,456,np.NaN,np.NaN]
                   }))
df['filled'] = df.code.ffill(axis = 0)
df