如何在具有多个值的列中选择部分字符串

Question

I have a.tsv Dataframe with a specific column with more than one value separated by commas. It looks like this:

     Col1         Col2         Col3

1    star1  HIP1, KOI1, Gaia1  3.4
2    star2  HIP2, KOI2, Gaia2  4.3
3    star3  HIP3, KOI3, Gaia3  7.2

My objective is to take only part of the string value from column 2, so that I only have one of the options separated by commas. In this case, it would be the KOIs. It'd look like this:

     Col1   Col2  Col3

1    star1  KOI1  3.4
2    star2  KOI2  4.3
3    star3  KOI3  7.2

Is there a way to do it, considering the numbers right after KOI(x) do not follow an ordinal order (as in the example)? I've tried using the str.lsplit() and split function, but the code returns the message: 'StringMethods' object has no attribute 'lsplit' . This is what I tried:

for i in df['Col2']:
    df['Col2'][i] = df['Col2'].str.lsptrip(', K').str[0]

I would then try adding the missing 'K' letter to the string, when I had it isolated, but never got to that part.

Answer 1

You could use pd.Series.str.extract too:

df['Col2']=df['Col2'].str.extract('.*, (K.*), .*')

---

Same as this, with pd.Series.str.split :

df['Col2']=df['Col2'].str.split(', ').str[1]

---

Output:

df
    Col1  Col2  Col3
1  star1  KOI1   3.4
2  star2  KOI2   4.3
3  star3  KOI3   7.2