Question

I can’t find a working regex in python to split these strings:

CAT One | desired: CAT

DOG SILVER FOX Two | desired: DOG SILVER FOX

KING KONG | desired: KING KONG

P'OT THEN Mark First | desired P'OT THEN

Just stupid examples, but i need to separate words that are full uppercase from words that are only capitalized.

I could have {1,n} uppercase words and {0,n} capitalized words.

My regexs were too weird, i catch all the string or only one uppercase word..

Answer 1

import re

lines = [
    "CAT One",
    "DOG SILVER FOX Two",
    " KING KONG ",
    "P'OT THEN Mark First",
    "FOO-BAR Second FISH",
    "horsE YELLOW thirD BLUE",
    ]

for line in lines:
    print re.findall(r'\b[A-Z]+(?:\W*[A-Z]+)*\b', line)

Output:

['CAT']
['DOG SILVER FOX']
['KING KONG']
["P'OT THEN"]
['FOO-BAR', 'FISH']
['YELLOW', 'BLUE']

Explanation:

\b[A-Z]+ means: match one or more capital letters, but only at the start of a word. This will match “YELLOW”, but not the “E” in “horsE”.

\W*[A-Z]+ means: match zero or more non-word characters, followed by one or more capital letters. This will match “‘OT” or “-BAR” or ” KONG”.

(?:\W*[A-Z]+)*\b means: make a (non-capturing) group which matches zero or more times, but only at the end of a word. This will match ” SILVER FOX”, but not the ” T” which follows it.