Best pythonic way to merge consecutive upper case characters in a string python

Question

I woud need assistance to find the best pythonic way to merge consecutive upper case characters in a string python

Example:

Input: You can pay N O W or Pay me Back MY Money later
Output: You can pay NOW or Pay me Back MY Money later

I am going with a very quick & dirty approach temporarily

s='lets P A Y N O W'

new_s = s
replace_maps = []
replace_str = ''
prev_cap = False

for i, c in enumerate(s):
    if c == ' ':
        continue
    if c.isupper():
        if prev_cap:
            replace_str += c
        else:
            start = i
            replace_str = c
            prev_cap = True
    else:
        end = i
        if prev_cap:
            replace_maps.append([start, end, replace_str])
        prev_cap = False
        replace_str = ''
else:
    end = i
    if prev_cap:
        replace_maps.append([start, end, replace_str])
    prev_cap = False
    replace_str = ''
    
new_s = s[:replace_maps[0][0]] + replace_maps[0][2] + s[replace_maps[0][1]:]
new_s

Output: lets PAYNOWW

Could you please share your attempt and specify where exactly you are stuck? — Cory Kramer, Sep 20 '22 at 12:31
@Cory Kramer, I am not stuck anywhere. just want the most pythonic approach. I'll update mine in a while — Yaser Sakkaf, Sep 20 '22 at 13:00

Lucas M. Uriarte · Accepted Answer · 2022-09-20T15:08:04.017

2

The best idea is to use Look-aheads ?= and Look-behinds ?<= and check for Upper case letters.

for more info on regex

this regex should make the job

import re

data = "I could not find a C O V I D patient in the hospital."
re.sub(r"(?<=[A-Z])\s(?=[A-Z])", r'', data)

'I could not find a COVID patient in the hospital.'

EDIT

Regarding your new input after question modification

data = "You can pay N O W or Pay me Back MY Money later"
re.sub(r"(?<=[A-Z])\s(?=[A-Z] )", r'', data)

output

'You can pay NOW or Pay me Back MY Money later'

edited Sep 20 '22 at 15:08

answered Sep 20 '22 at 13:34

Lucas M. Uriarte

2,403
5
19

1

Oh, amazing and elegant way! – Carapace Sep 20 '22 at 13:53
Hi @Lucas M. Uriarte, please see the question again. I've change the Input Example which doesn't get solved by your approach – Yaser Sakkaf Sep 20 '22 at 14:20
@YaserSakkaf. Done. You only need to add a white space to skip words – Lucas M. Uriarte Sep 20 '22 at 15:09
Thanks @Lucas M. Uriarte, this solved it – Yaser Sakkaf Sep 22 '22 at 12:16

score 1 · Answer 2 · answered Sep 20 '22 at 14:03

1

without regex:

mystring = mystring.split(" ")
res = mystring[0]
for i in range(1, len(mystring)):
   if not (mystring[i-1].isupper() and mystring[i].isupper()):
      res+= " "
   res += mystring[i]

answered Sep 20 '22 at 14:03

Axeltherabbit

680
3
20

score 0 · Answer 3 · answered Sep 20 '22 at 13:12

I don't know what the most pythonic way could be. I can only tell you what I came up.

import re


def merge_cons_up(string):
    pattern = re.compile(" [A-Z](?![a-zA-Z0-9_.-])")
    sub_text = re.findall(pattern=pattern, string=string)
    repl = "".join(sub_text).replace(" ", "")
    sub = re.sub(pattern=pattern, string=string, repl=" " + repl, count=1)
    final_string = re.sub(pattern=pattern, string=sub, repl="")
    return final_string


print(merge_cons_up("I could not find a C O V I D patient in the hospital."))

Output:

I could not find a COVID patient in the hospital.

Best pythonic way to merge consecutive upper case characters in a string python

3 Answers3