If condition on one column if true then find pattern on another column and give output pandas

Question

Below is the data I am working on:

df_raw = pd.DataFrame({'Summary':['|ro-rd4_ae20|Issue-backfgound', '|20:36|site1_shutdown'], 'User':[r'UPC\User',r'UPC\Ankita'], 'Name':['Generic User', 'CSD']})

Using a regular expression, I want to check pattern for 'Name' =CSD and 'Name' ='Generic User' separately which will create new column giving true / false value if matches pattern in re.

If df_raw.Name ='CSD' then apply reg. exp (df_raw['Summary'].str.findall(r'(([?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z-._]+)', expand=False))

and df_raw.Name = 'Generic User' then apply reg exp (df_raw['Summary'].str.findall(r'(([?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z-._]+)', expand=False))

I have tried adding re in variable and apply but that is not helping/not giving output

pls help with this

yes @HenryYik their will be some change in pattrn which i will do later — Trupti Chaudhari, Oct 31 '20 at 14:54
@NickODell bcz the requirement to match the pattern here and I thought re is the option I can go for — Trupti Chaudhari, Oct 31 '20 at 17:14
It looks like it is the problem related to the question rather than the answer. TruptiChaudhari, look, you have two identical regexps to apply in both cases. Please review your question, make it clear and state the exact required output. — Wiktor Stribiżew, Nov 01 '20 at 14:32

Wiktor Stribiżew · Accepted Answer · 2020-11-01T15:57:50.473

You can use

import pandas as pd
import re

df_raw = pd.DataFrame({'Summary':['|ro-rd4_ae20|Issue-backfgound', '|20:36|site1_shutdown'], 'User':[r'UPC\User',r'UPC\Ankita'], 'Name':['Generic User', 'CSD']})

def extract(r):
  if r["Name"] == "Generic User":
    return bool( re.search(r'(?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z._-]+', r["Summary"]) )
  elif r["Name"] == "CSD":
    return bool( re.search(r'(?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z._-]+', r["Summary"]) )
  return False

df_raw["Valid"] = df_raw.apply(extract, axis=1)

Output:

>>> df_raw
                         Summary        User          Name  Valid
0  |ro-rd4_ae20|Issue-backfgound    UPC\User  Generic User   True
1          |20:36|site1_shutdown  UPC\Ankita           CSD   True

The df_raw["Valid"] column will contain the True or False values.

Note I removed a "wild" [ at the start of your patterns, it seems off.

If you need to extract the matches use

def extract(r):
  if r["Name"] == "Generic User":
    m = re.search(r'(?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z._-]+', r["Summary"])
    if m: return m.group()
  elif r["Name"] == "CSD":
    m = re.search(r'(?:[01]?\d|2[0-9]):[0-9]\d|[a-z0-9A-Z._-]+', r["Summary"])
    if m: return m.group()
  return ''

>>> df_raw["Valid"] = df_raw.apply(extract, axis=1)
>>> df_raw
                         Summary        User          Name        Valid
0  |ro-rd4_ae20|Issue-backfgound    UPC\User  Generic User  ro-rd4_ae20
1          |20:36|site1_shutdown  UPC\Ankita           CSD        20:36
>>>

pls also help in how can I return int insted of bool – Trupti Chaudhari Nov 01 '20 at 08:20 — Trupti Chaudhari, Nov 01 '20 at 08:20

If condition on one column if true then find pattern on another column and give output pandas

1 Answers1