Is there an option to extract a pattern from strings in pandas?

i have to extract histologie codes from strings in pandas.

For example:

Histologie (ICD-O-2): 8500/2.
Histologie (ICD-O-3): 8842/8
(ICD-O-2): 8522/2. Histo

To:

8500/2
8842/8
8522/2

There are many variations of the writing in the original format.
For that reason i want to search the strings for the composition of digit,digit,digit,digit,char,digit (0000/0).

Thanks everyone for your help.

>Solution :

Assuming your columns is named "col", you can use a simple regex:

df['code'] = df['col'].str.extract(r'(\d{4}/\d)')

output:

                             col    code
0  Histologie (ICD-O-2): 8500/2.  8500/2
1   Histologie (ICD-O-3): 8842/8  8842/8
2       (ICD-O-2): 8522/2. Histo  8522/2

regex:

\d{4}  # match 4 digits
/      # match a literal /
\d     # match one digit

If you need to ensure that the codes are independent words (e.g., 12345/6a shouldn’t match:

df['code'] = df['col'].str.extract(r'(\b\d{4}/\d\b)')

or if non-digits are allowed to touch the code:

df['code'] = df['col'].str.extract(r'(?:\D|^)(\d{4}/\d\b)(?:\D|$)')