CODEWITHSUNDEEP
pythonpandasdataframe
Max
Max

Reputation: 471

How to make conditional formatting work in pandas?

I have the following df:

Invoice number
.5
.3
02

I want to replace the '.' because it gets captured incorrectly. This is succesful with the following code:

df['Invoice number'] = df['Invoice number'].astype(str).str.replace(r"[^a-zA-Z0-9\$?!]", '')

however, a 0 is placed in front of the number resulting in the following df:

Invoice number
05
03
02

I've tried the following

for i, row in df.iterrows():
    if str(i).startswith('.'):  
        df.loc[i, 'KvK'] = df['KvK'].astype(str).str.replace(r"[^a-zA-Z0-9\$]", '')
        df.loc[i,'Factuurnummer'] = df['Factuurnummer'].astype(str).str.replace(r"[^a-zA-Z0-9\$]", '')

However, this does not work. Nothing happens.

I would like the following output:

invoice # 
5
3
02

Upvotes: 2

Views: 131

Answers (1)

anky
anky

Reputation: 75080

you can do this with a mask created by series.str.startswith and series.mask or np.where:

s = df['Invoice number']
s.mask(s.str.startswith('.'),s.str.replace(".",""))

Or if one is comfortable using a regex pattern, one can also use:

df['Invoice number'].str.replace(r'^\.', '')

0     5
1     3
2    02
Name: Invoice number, dtype: object

Upvotes: 2

Related Questions