Using a regular expression to replace upper case repeated letters in python with a single lowercase letter

Question

I am trying to replace any instances of uppercase letters that repeat themselves twice in a string with a single instance of that letter in a lower case. I am using the following regular expression and it is able to match the repeated upper case letters, but I am unsure as how to make the letter that is being replaced lower case.

import re
s = 'start TT end'
re.sub(r'([A-Z]){2}', r"1円", s)
>>> 'start T end'

How can I make the "1円" lower case? Should I not be using a regular expression to do this?

• khachik: Don't know how to make it lowercase, but your should use '([A-Z]){2,}' instead of '([A-Z]){2}' to replace any instances.
• Sven Marnach: Your regex also matches two different caps.

Answer 1

Pass a function as the repl argument. The MatchObject is passed to this function and .group(1) gives the first parenthesized subgroup:

import re
s = 'start TT end'
callback = lambda pat: pat.group(1).lower()
re.sub(r'([A-Z]){2}', callback, s)

EDIT
And yes, you should use ([A-Z])1円 instead of ([A-Z]){2} in order to not match e.g. AZ. (See @bobince's answer.)

import re
s = 'start TT end'
re.sub(r'([A-Z])1円', lambda pat: pat.group(1).lower(), s) # Inline

Gives:

'start t end'

Answer 2

You can't change case in a replacement string. You would need a replacement function:

>>> def replacement(match):
... return match.group(1).lower()
... 
>>> re.sub(r'([A-Z])1円', replacement, 'start TT end')
'start t end'

Answer 3

def replace(s):
 return " ".join(re.findall(r"[A-Z]){2}", s)).lower()

I guess this is what you are looking for.

Answer 4

You can do it with a regular expression, just pass a function as the replacement like the docs say. The problem is your pattern.

As it is, your pattern matches runs of any two capital letters. I'll leave the actual pattern to you, but it starts with AA|BB|CC|.

Answer 5

The 'repl' parameter that identifies the replacement can be either a string (as you have it here) or a function. This will do what you wish:

import re
def toLowercase(matchobj):
 return matchobj.group(1).lower()
s = 'start TT end'
re.sub(r'([A-Z]){2}', toLowercase, s)
>>> 'start t end'

Answer 6

Try this:

def tol(m):
 return m.group(0)[0].lower()
s = 'start TTT AAA end'
re.sub(r'([A-Z]){2,}', tol, s)

Note that this doesn't replace singe upper letters. If you want to do it, use r'([A-Z]){1,}'.

• SilentGhost: OP says: repeat themselves twice
• khachik: @SilentGhost. My fault. the re should be as suggested by Ignacio, if single upper characters shouldn't be touched.
• SilentGhost: if you look and bobince's and jens's answers you see the shorter way to do this.
• khachik: I see, thanks. The question about single upper chars is still open.

Answer 7

WARNING! This post has no re as requested. Continue with your own responsibility!

I do not know how possible are corner cases but this is how normal Python does my naive coding.

import string
s = 'start TT end AAA BBBBBBB'
for c in string.uppercase:
 s = s.replace(c+c,c.lower())
print s
""" Output:
start t end aA bbbB
"""