Replacing repetitive words, case-insensitively

Question

Sample:

>>> line = 'the the, To to'

>>> re.findall(r'\b(\w+) \1', line)

['the']

>>> re.findall(r'\b(\w+) \1', line, re.I)

['the', 'To']



>>> re.sub(r'\b(\w+) \1', r'\1', line, re.I)

'the, To to'

Expected:

'the, To'

The regex works in other places like

Vim:
s/\v<(\w+) \1/\1/gi

Perl:
s/\b(\w+) \1/$1/gi

sed:
-r 's/\b(\w+) \1/\1/gi'

Is this a known behavior? What is a workaround? My Python version is

3.4.3

if that makes a difference.

Recommended for you: Get network issues from WhatsUp Gold. Not end users.

Answer 1

Read the definition of re.sub:

re.sub(pattern, repl, string, count=0, flags=0)

You are passing re.I as count (where it is allowing at most 2 replacements), not as flags. Instead, try:

>>> re.sub(r'\b(\w+) \1', r'\1', s, flags=re.I)
                                  # ^ note
'the, To'

Replacing repetitive words, case-insensitively

Sign up