Combining two unique values in a Pandas column if they have the same value in another column

Let's say I have a very large Pandas dataframe in Python that looks something like this:

df_test = pd.DataFrame(data = None, columns = ['file','source'])
df_test.file = ['file_1', 'file_1', 'file_2', 'file_2', 'file_3', 'file_3']
df_test.source = ['usa', 'uk', 'jp', 'sk', 'au', 'nz']

What I want to get out from this is for the 'source' column to combine the unique sources into a single string separating the two unique sources with a '; ' for each value in the 'file' column that is the same. The end result for the 'source' column should therefore be:

['usa; uk', 'usa; uk', 'jp; sk', 'jp; sk', 'au; nz', 'au; nz']

Since 'file_1' in the 'file' column has the two sources 'usa' and 'uk', etc. The actual dataframe is very large so it must be done automatically and not manually. Any help on how to do this would be really appreciated, thanks!

Use lambda function in GroupBy.transform with remove duplicated values in dict.fromkeys or by sets:

df_test['new'] = (df_test.groupby('file')['source']
                         .transform(lambda x: '; '.join(dict.fromkeys(x))))
print(df_test)
     file source      new
0  file_1    usa  usa; uk
1  file_1     uk  usa; uk
2  file_2     jp   jp; sk
3  file_2     sk   jp; sk
4  file_3     au   au; nz
5  file_3     nz   au; nz

df_test['new'] = df_test.groupby('file')['source'].transform(lambda x: '; '.join(set(x)))
print(df_test)
     file source      new
0  file_1    usa  uk; usa
1  file_1     uk  uk; usa
2  file_2     jp   jp; sk
3  file_2     sk   jp; sk
4  file_3     au   nz; au
5  file_3     nz   nz; au