Regex pattern used:
(?<!cart .*)(?<!order .)(\b\d{4,12}\b)
Examples:
so the cart number is 12344567 that it
my cart number is 1234
order number note down 12345
order 12345
credit card is 0000
4444
the 5456
Expected output:
0000
4444
5456
Here i don't want number that starts with word order or cart that can range from 4 to 12.
If you want to use Python with a infinite quantifier in the lookbehind, you can use the PyPi regex module.
You don't need the capture group, and you have to repeat the dot for the order part (?<!order .*)
But using an alternation, you can use a single lookbehind:
(?<!(?:cart|order) .*)\b\d{4,12}\b
See a regex demo.
Example
import regex
pattern = r"(?<!(?:cart|order) .*)\b\d{4,12}\b"
s = ("so the cart number is 12344567 that it\n\n"
"my cart number is 1234\n\n"
"order number note down 12345\n\n"
"order 12345\n\n"
"credit card is 0000\n\n"
"4444\n\n"
"the 5456")
print(regex.findall(pattern, s))
Output
['0000', '4444', '5456']
Another option is to use an alternation with re.findall and filter out the empty values, or use re.finditer and loop the values checking for a group 1 value.
An example with re.findall, where you match what you don't want and capture in group 1 what you want to keep using a for comprehension:
import re
pattern = r"^.*\b(?:cart|order)\b.*|(\b\d{4,12}\b)"
s = ("so the cart number is 12344567 that it\n"
"my cart number is 1234\n"
"order number note down 12345\n"
"order 12345\n"
"credit card is 0000\n"
"4444\n"
"the 5456")
print([s for s in re.findall(pattern, s, re.M) if s])
Output
['0000', '4444', '5456']
re