python pandas data frame create columns if list of string contained in one columns-CodePudding

given this df:

data = {'Description':  ['with milk and orange', 'champagne', 'BANANA', 'bananas and apple', 'fafsa Lemons', 'GIN LEMON'],
        'Amount': ['10', '20', '10', '5', '9', '15']}
df = pd.DataFrame(data)
print (df)

and the following vector:

Fruits = ['apple','banana','lemon', 'orange']

how to obtain the column 'Fruit'? (search for all the elements of the vector Fruits in the column description and add them in the Column 'fruit' if contained in Description)

datanew = {'Description':  ['with milk and orange', 'champagne', 'BANANA', 'bananas and apple', 'fafsa Lemons', 'GIN LEMON'],
        'Amount': ['10', '20', '10', '5', '9', '15'],
        'Fruit':  ['orange', '', 'banana', 'banana-apple', 'lemon', 'lemon'],
       
       }
df2 = pd.DataFrame(datanew)
print (df2)

CodePudding user response：

You can use str.extractall and groupby.agg:

import re
df['Fruit'] = (df['Description']
               .str.extractall(f"({'|'.join(Fruits)})", flags=re.I)
               .groupby(level=0).agg('-'.join)[0]
               .str.lower()
              )

output:

            Description Amount         Fruit
0  with milk and orange     10        orange
1             champagne     20           NaN
2                BANANA     10        banana
3     bananas and apple      5  banana-apple
4          fafsa Lemons      9         lemon
5             GIN LEMON     15         lemon

CodePudding user response：

Here is another way by using str.findall()

(df.assign(Fruit = df['Description'].str.lower()
.str.findall('|'.join(Fruits))
.str.join('-')))

            Description Amount         Fruit
0  with milk and orange     10        orange
1             champagne     20              
2                BANANA     10        banana
3     bananas and apple      5  banana-apple
4          fafsa Lemons      9         lemon
5             GIN LEMON     15         lemon