Missing results after loop Python-CodePudding

I have the above dataframes..

df_a1 = pd.DataFrame([['ab1', 10], ['ab2', 15], ['ab3', 14]], columns=['listing_id', 'number_a1'])
df_a2 = pd.DataFrame([['aabb1', 5], ['aabb2', 6], ['aabb3', 7]], columns=['listing_id', 'number_b1'])

df_b1 = pd.DataFrame([['ab1', 20], ['ab3', 25], ['ab4', 20]], columns=['id', 'number_b1'])
df_b2 = pd.DataFrame([['aabb3', 13], ['aabb2', 12], ['aabb4', 11]], columns=['id', 'number_b2'])

I need to merge with left_outer the df_a1 with df_b1 and df_a2 with df_b2. Because in reality they are too many data frames I need an iteration.

I have try this..

df_full_aa = pd.DataFrame()
df_full_bb = pd.DataFrame()

df_list = [df_full_aa, df_full_aa]
df_aa = [df_a1, df_a2]
df_bb = [df_b1, df_b2]  

for i in range(len(df_list)):
    df_list[i] = pd.merge(df_aa[i], df_bb[i], how="left", left_on='listing_id', right_on='id')
    print([str(i)])
    print(df_list[i].head())

The print inside the loop looks like it work but if i try to print the dfs outside the results are missing.

After many search i have tried also this..

for i in enumerate(df_list):
    df_list[i] = pd.merge(df_aa[i], df_bb[i], how="left", left_on='listing_id', right_on='id')

but it returns the above error

TypeError: list indices must be integers or slices, not tuple

I have already try to understand this Python loops are missing results but i can't adjust it to my case.

Thank you for your time!!!

CodePudding user response：

df_full_aa df_full_bb are empty, because these variables still point to the same empty dataframes that you have created at the beginning. The new dataframes are stored in your df_list: you can try this after the loop: df_full_aa = df_list[0] and df_full_bb = df_list[1].

You don't actually need to create empty df_full_aa and df_full_bb before the loop. Here is a better version:

df_list = []
df_aa = [df_a1, df_a2]
df_bb = [df_b1, df_b2]  

for i in range(len(df_aa)):
    df_list.append(pd.merge(df_aa[i], df_bb[i], how="left", left_on='listing_id', right_on='id'))
    print([str(i)])
    print(df_list[i].head())

df_full_aa, df_full_bb = df_list