Pandas, merge two dataframes based on one column that contains the other

Question

I have the following dataframe B:

|index|cond2                                                |
|-----|-----------------------------------------------------|
|0    |            ((age <= 23), ((age <= 31) & (23 < age)))|
|1    |                         (((age <= 38) & (31 < age)))|
|2    |    (((age <= 50) & (43 < age)), ((age <= 43) & (3...|
|3    |                         (((age <= 54) & (50 < age)))|
|4    |                         (((age <= 64) & (54 < age)))|
|5    |                                         ((64 < age))|

which contains thousands of lines. Note that cond2 values are strings.

the column cond1 (its values are also strings) was created from the following table A:

|cond1                   |note|
|------------------------|----|
|(age <= 23)             |20  |
|(age <= 31) & (23 < age)|30  |
|(age <= 38) & (31 < age)|40  |

How may I merge both tables to add the column 'note' to table B if 'cond1' is contained in 'cond2'. The expected output is the following table:

index	cond2	cond1	note
0	((age <= 23), ((age <= 31) & (23 < age)))	(age <= 23)	20
1	(((age <= 38) & (31 < age)))	((age <= 38) & (31 < age))	40
2	(((age <= 50) & (43 < age)), ((age <= 43) & (3...	((age <= 50) & (43 < age))	60

As far, I have tried to used the function I found in a similar question (Python Pandas DataFrame str contains merge if) but unfortunately, It doesn't work in this specific case:

def str_merge(part_string_df,full_string_df, merge_column):
    merge_column_lower = 'merge_column_lower'
    part_string_df[merge_column_lower] = part_string_df[merge_column].str.lower()
    full_string_df[merge_column_lower] = full_string_df[merge_column].str.lower()   
    pat = '|'.join(r"{}".format(x) for x in part_string_df[merge_column_lower])
    full_string_df['Test3'] = full_string_df[merge_column_lower].str.extract('('+ pat + ')', expand=True)
    DF = pd.merge(part_string_df, full_string_df, left_on= merge_column_lower, right_on='Test3').drop([merge_column_lower + '_x',merge_column_lower + '_y','Test3'],axis=1)
    return DF

Answer 1

Try:

dfB["cond1"] = dfB["cond2"].apply(lambda x: [c for c in dfA["cond1"] if c in x])
dfB = dfB.explode("cond1").drop_duplicates("cond2")
output = dfB.merge(dfA)

>>> output
                                               cond2  ... note
0          ((age <= 23), ((age <= 31) & (23 < age)))  ...   20
1                       (((age <= 38) & (31 < age)))  ...   40
2  (((age <= 50) & (43 < age)), ((age <= 43) & (3...  ...   60