Repository navigation
re.Scanner groups #84440
Description
| BPO | 40259 |
|---|---|
| Nosy | @ezio-melotti, @tirkarthi |
| Files |
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 = None
created_at = <Date 2020-04-12.07:52:00.286>
labels = ['expert-regex', 'type-bug', '3.8']
title = 're.Scanner groups'
updated_at = <Date 2020-04-12.14:06:10.839>
user = 'https://bugs.python.org/dchron'bugs.python.org fields:
activity = <Date 2020-04-12.14:06:10.839>
actor = 'xtreak'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Regular Expressions']
creation = <Date 2020-04-12.07:52:00.286>
creator = 'dchron'
dependencies = []
files = ['49049']
hgrepos = []
issue_num = 40259
keywords = []
message_count = 2.0
messages = ['366226', '366249']
nosy_count = 4.0
nosy_names = ['ezio.melotti', 'mrabarnett', 'xtreak', 'dchron']
pr_nums = []
priority = 'normal'
resolution = None
stage = None
status = 'open'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue40259'
versions = ['Python 3.8']
Activity
tirkarthi commentedon Apr 12, 2020
Please add a description of the issue you are facing with a simple script of the behavior.
tirkarthi commentedon Apr 12, 2020
Copy paste of the contents in the text file
In the re module there is an experimental feature called Scanner.
Some unexpected behavior was found while working with it.
Here is an example:
The obvious error is the semicolon returned via capturing group 1.
Adding a dummy rule at the beginning, seems to solve that issue:
Adding a capturing group around \w+ also returns the correct answer:
But then, if I ask for the second group, the problem appears again:
serhiy-storchaka commentedon Apr 18, 2022
re.Scanneruses groups internally. You should not refer to groups by number unless you know exactly howre.Scannerworks internally. Use named groups instead.It could be a documentation issue, but
re.Scanneris not officially documented, so nothing to do here.serhiy-storchaka commentedon Oct 31, 2025
My suggestion to use named groups was incorrect -- no capturing groups are supported in
re.Scanner(). Even named group change numbering of groups and mess with capturing groups used byre.Scanner()itself.