From a df column 'Desc' I want to extract substrings that start with n or N followed by a digit, here's a test df with my code and result:
import pandas as pd
testdf = pd.DataFrame({'Desc': ['n1.2A Full Version', 'N5.0.0 Bridge', 'N5.35A Automatic', 'n2 Bridge']})
testdf['Version'] = testdf['Desc'].str.extract(r'([nN]\d.+?[\s])', expand=False)
How to fix the regex so that it doesn't show NaN for the last record? Thanks
The main issue is that .+? requires at least 1 char other than line break char and then [\s] requires a whitespace to match. So, when [nN]\d matched and consumed n2, the regex engine tries to match the next space with .+? and then [\s] fails to match a whitespace.
You may use
>>> testdf['Desc'].str.extract(r'([nN]\d\S*)', expand=False)
0 n1.2A
1 N5.0.0
2 N5.35A
3 n2
The pattern is [nN]\d\S*:
[nN] - n or N\d - a digit\S* - 0 or more non-whitespace charsSee online regex demo and the regex graph:
