➜
This issue tracker has been migrated to GitHub,
and is currently read-only.
For more information,
see the GitHub FAQs in the Python's Developer Guide.
Created on 2015-01-22 15:58 by xflr6, last changed 2022-04-11 14:58 by admin. This issue is now closed.
| Messages (2) | |||
|---|---|---|---|
| msg234497 - (view) | Author: Sebastian Bank (xflr6) | Date: 2015-01-22 15:58 | |
ConfigParser parses section lines containing square brackets like '[spam [eggs] spam]' up to the first instead of the last occurrence of ']' preventing roundtrips:
>>> s = StringIO()
>>> c1 = ConfigParser()
>>> c1.add_section('spam [eggs]')
>>> c1.write(s)
>>> s.seek(0)
>>> c2 = ConfigParser()
>>> c2.readfp(s)
>>> assert c1.sections() == c2.sections() # fails
Potential fix: change the second line of SECTCRE from r'(?P<header>[^]]+)' to r'(?P<header>.+?)'.
If the parsing behaviour cannot be changed, the user should at least be warned about supplying data that breaks the roundtrip.
|
|||
| msg234498 - (view) | Author: R. David Murray (r.david.murray) *  |
Date: 2015-01-22 16:06 | |
This is a duplicate of issue 20923, which was rejected. To argue against the rejection you probably need to provide evidence that this is something that is actually supported by other common ini parsers. And that evidence should be posted to issue 20923. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-11 14:58:12 | admin | set | github: 67490 |
| 2015-01-22 16:06:52 | r.david.murray | set | status: open -> closed superseder: [doc] Explain ConfigParser 'valid section name' and .SECTCRE nosy: + r.david.murray messages: + msg234498 resolution: duplicate stage: resolved |
| 2015-01-22 15:58:05 | xflr6 | create | |