I have a concatenated text that I want to split using Regex. Luckily there is a pattern. The pattern is structured this way: (seconds) some text (seconds) some other text (seconds) some other text

(1-4) Agent: THANK YOU FOR CALLING XCOMPANY MY NAME IS DEVIN HOW CAN I HELP YOU (5-22) Customer: HI KEVIN I TRANSFERRED OVER TO YOU ... (24-29) Agent: OK AY (662-662) Customer: THANK YOU TOO (663-664) Agent: THANKS BYE NOW (664-664) Customer: BYE

I want to split each block so output should be like this.

(1-4) Agent: THANK YOU FOR CALLING XCOMPANY MY NAME IS DEVIN  HOW CAN I HELP YOU 
(5-22) Customer: HI KEVIN I TRANSFERRED OVER TO YOU ... ABOUT THAT BILL 
(24-29) Agent: OK  AY 

So far I was able to create this \(\d*-\d*\)\s*\w*:\s*, but this catches (1-4) Agent: I can't figure out the rest, I tried many things.

Here is Regex101 link, showing where I am stuck.

CodePudding user response：

you could try Match groups

(\(\d*-\d*\)\s*\w :[\s?\w] [^(]){1}

I'm not much of a regex pro but I do try. Let me know if it helped :D

CodePudding user response：

With

\(\d*-\d*\)\s*\w*:[^(]*

you can catch everything after the colon that is not an open parenthesis.

CodePudding user response：

In the pattern that you have tried, the digits between parenthesis are optional due to the *, and the \w*:\s* does not match beyond optional word characters : and optional whitespace chars.

---

You can use:

\(\d -\d \).*?(?=\(\d -\d \)|$)

Explanation

• \(\d -\d \) match (, 1 digits - 1 digits and )
• .*? Match any character, as few as possible
• (?= Positive lookahead
  • \(\d -\d \) The digit pattern between parenthesis
  • | Or
  • $ End of string (For the last occurrence)
• ) Close lookahead

Regex demo

Example code

import re

pattern = r"\(\d -\d \).*?(?=\(\d -\d \)|$)"

s = "(1-4) Agent: THANK YOU FOR CALLING XCOMPANY MY NAME IS DEVIN HOW CAN I HELP YOU (5-22) Customer: HI KEVIN I TRANSFERRED OVER TO YOU ... (24-29) Agent: OK AY (662-662) Customer: THANK YOU TOO (663-664) Agent: THANKS BYE NOW (664-664) Customer: BYE"

print(re.findall(pattern, s))

Output

[
'(1-4) Agent: THANK YOU FOR CALLING XCOMPANY MY NAME IS DEVIN HOW CAN I HELP YOU ',
'(5-22) Customer: HI KEVIN I TRANSFERRED OVER TO YOU ... ',
'(24-29) Agent: OK AY ',
'(662-662) Customer: THANK YOU TOO ',
'(663-664) Agent: THANKS BYE NOW ', '(664-664) Customer: BYE'
]