using substring and if elif condition to create new column

[ck25python]

Hi There,
I am new to python so please be kind to me.
Below is the data frame and the requirement:

data = {'id': ['aa11bc', 'bb22cd', 'cc33ef', 'dd44gh', 'ee55ij','ff66kl','gg77mn','hh88op'], 
        'direction': ["north, south, east, west", "north, south, east, west", "north, south/, *east, \west%", "north, south, east, west",
                      "north, south, east, west","north, south, east, west","north, south, east, west"
                     ,"north, south, east, west"]}
df = pd.DataFrame(data, columns = ['id','direction'])
df

Requirement:
#if id (2nd and 3rd letter) in ('a1','b2') then new_col should have north as an observation from direction column
#else if id (2nd and 3rd letter) in ('c3','d4') then new_col should have south as an observation from direction column
#else if id (2nd and 3rd letter) in ('e5','f6') then new_col should have east as an observation from direction column
#else if id (2nd and 3rd letter) in ('g7','h8') then new_col should have west as an observation from direction column

I tried with the below code: but the output is not correct it's not meeting the desired output.

new_col=[]
 
for i in df["id"]:
    if i[1:3].lower()in ('a1','b2'):
        new_col=df["direction"].str.split(',').str[0].str.replace('\W+',' ').str.strip()
    elif i[1:3].lower()in ('c3','d4'):
        new_col=df["direction"].str.split(',').str[1].str.replace('\W+',' ').str.strip()
    elif i[1:3].lower()in ('e5','f6'):
        new_col=df["direction"].str.split(',').str[2].str.replace('\W+',' ').str.strip()
    elif i[1:3].lower()in ('g7','h8'):
        new_col=df["direction"].str.split(',').str[3].str.replace('\W+',' ').str.strip()
         
df["position"]=new_col
print(df)

And the output as follows:
       id                     direction position
0  aa11bc      north, south, east, west     west
1  bb22cd      north, south, east, west     west
2  cc33ef  north, south/, *east, \west%     west
3  dd44gh      north, south, east, west     west
4  ee55ij      north, south, east, west     west
5  ff66kl      north, south, east, west     west
6  gg77mn      north, south, east, west     west
7  hh88op      north, south, east, west     west

Please advise.

[SioSio]

I didn't understand what you wanted to do, but is that OK?
Do you need to replace non-alphabets in direction?

new_col=[]
j = 0
for i in df["id"]:
    if i[1:3].lower()in ('a1','b2'):
        new_col.append(re.sub(r'\W+','',df.loc[j, 'direction'].split(',')[0]))
    elif i[1:3].lower()in ('c3','d4'):
        new_col.append(re.sub(r'\W+','',df.loc[j, 'direction'].split(',')[1]))
    elif i[1:3].lower()in ('e5','f6'):
        new_col.append(re.sub(r'\W+','',df.loc[j, 'direction'].split(',')[2]))
    elif i[1:3].lower()in ('g7','h8'):
        new_col.append(re.sub(r'\W+','',df.loc[j, 'direction'].split(',')[3]))
    j = j + 1
df ["position"] = new_col
print(df)

[ck25python]

Hi There,
Thanks for the help.
Yes, I am getting the desired output from your code.
I want to try one more condition from your code, I let you know the status. Until such time I will keep this question as open.
Hope this is fine with you.
Once again thanks for your help.
Cheers!!

[ck25python]

HI SioSio,
Thanks for the help with the code.