re.sub() bug with IGNORECASE

[DanWilt]

BPO	27851
Nosy	@ezio-melotti, @bitdancer
Files	re_bug.py: MWE for re bug

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2016-08-24.15:59:54.318>
created_at = <Date 2016-08-24.15:57:11.028>
labels = ['expert-regex', 'type-bug', 'invalid']
title = 're.sub() bug with IGNORECASE'
updated_at = <Date 2016-08-24.18:28:02.106>
user = 'https://bugs.python.org/DanWilt'

bugs.python.org fields:

activity = <Date 2016-08-24.18:28:02.106>
actor = 'Dan Wilt'
assignee = 'none'
closed = True
closed_date = <Date 2016-08-24.15:59:54.318>
closer = 'r.david.murray'
components = ['Regular Expressions']
creation = <Date 2016-08-24.15:57:11.028>
creator = 'Dan Wilt'
dependencies = []
files = ['44212']
hgrepos = []
issue_num = 27851
keywords = []
message_count = 3.0
messages = ['273575', '273576', '273584']
nosy_count = 4.0
nosy_names = ['ezio.melotti', 'mrabarnett', 'r.david.murray', 'Dan Wilt']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue27851'
versions = ['Python 2.7', 'Python 3.5']

[DanWilt]

Working with re.sub() noted strange behavior with re.I set, seems like a bug. Noted in both Python 2.7.12 and Python 3.5.2, Anaconda custom build (32-bit) on Windows 7.

>>> import re
>>> re.sub('\.', '', '.....')
''
>>> re.sub('\.', '', '.....', re.I)
'...'
>>> re.sub('\.', '', '.....', re.L)
'.'

The first case is the expected behavior, the second and third aren't. Oddly enough,

>>> re.sub('\.', '', '.....', re.L | re.I)
''

MWE file attached (python3)

Thanks, I'm a heavy python user, but not all that advanced - hope this is helpful and I'm not just being stupid.

[bitdancer]

I really wish we could "fix" this somehow, since we get an issue opened for it probably every two or three months.

The fourth argument to re.sub is the count, not the flags.

[DanWilt]

Thanks! That helps. I apologize for raising a non-existent issue.
Dan

On Wed, Aug 24, 2016 at 11:59 AM, R. David Murray <[email protected]>
wrote:

R. David Murray added the comment:

I really wish we could "fix" this somehow, since we get an issue opened
for it probably every two or three months.

The fourth argument to re.sub is the count, not the flags.

----------
nosy: +r.david.murray
resolution: -> not a bug
stage: -> resolved
status: open -> closed

---

Python tracker <[email protected]>
<https://bugs.python.org/issue27851\>

---