yogi_peace yogi_peace - 6 months ago 16
Python Question

Pandas- Split text in column and search in rows

This question is with this history: Link

Here is a json format table:

ID Title
19 I am doing great
25 [Must fix problem] Stomach not well
31 [Not-so-urgent] Wash cloths
498 [VERY URGENT] Pay your rent
517 Landlord wants you to pay your rent tomorrow
918 Girlfriend wants to help you to pay rent if you take her out
1000 [Always reproducible issue] Room partner dont want to pay any rent, he is out of cash


I did this

In:
selected_row_title = df.loc[df['id'] == 4]["title"]

Output:

[VERY URGENT] Pay your rent


Now, by using Python Pandas, I am trying to write a function as:

get_matching_rows(selected_row_title )


Output

ID 498 has pay your rent
ID 517 has pay your rent
ID 918 has pay rent
ID 1000 has pay rent


I have been tearing my hair out on this and I really need some help, atleast a guidance on how once can implement this. Appreciate any inputs.

Answer

I think you can use str.replace with str.contains:

s = "[VERY URGENT] Pay your rent"

#replace all [] in column title
tit = df.Title.str.replace(r'[\[\]]', '')
print (tit)

0                                     I am doing great
1                    Must fix problem Stomach not well
2                            Not-so-urgent Wash cloths
3                            VERY URGENT Pay your rent
4         Landlord wants you to pay your rent tomorrow
5    Girlfriend wants to help you to pay rent if yo...
6    Always reproducible issue Room partner dont wa...
Name: Title, dtype: object

#search one of word of string s (logical or is |)
mask = tit.str.contains(s.replace(' ', '|'))
print (mask)
0    False
1    False
2     True
3     True
4     True
5     True
6     True
Name: Title, dtype: bool
#select all ID by condition
selected_row_title = df.loc[mask, 'ID']
print (selected_row_title)
2      31
3     498
4     517
5     918
6    1000
Name: ID, dtype: int64