No deprecation warning for list comprehension leak conflict

[mrbax]

BPO	6580
Nosy	@terryjreedy

^{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 2010-08-03.21:31:39.458>
created_at = <Date 2009-07-26.17:25:01.903>
labels = ['interpreter-core', 'type-bug', 'invalid']
title = 'No deprecation warning for list comprehension leak conflict'
updated_at = <Date 2010-08-03.21:31:39.457>
user = 'https://bugs.python.org/mrbax'

bugs.python.org fields:

activity = <Date 2010-08-03.21:31:39.457>
actor = 'terry.reedy'
assignee = 'none'
closed = True
closed_date = <Date 2010-08-03.21:31:39.458>
closer = 'terry.reedy'
components = ['Interpreter Core']
creation = <Date 2009-07-26.17:25:01.903>
creator = 'mrbax'
dependencies = []
files = []
hgrepos = []
issue_num = 6580
keywords = []
message_count = 3.0
messages = ['90952', '91150', '112690']
nosy_count = 2.0
nosy_names = ['terry.reedy', 'mrbax']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = None
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue6580'
versions = ['Python 2.5']

[mrbax]

PEP-289 states that "Python 2.4 and beyond should issue a deprecation
warning if a list comprehension's loop variable has the same name as a
variable used in the immediately surrounding scope". But no warning is
shown when running
for x in [ 0 ]: print [ x for x in [ 1 ] ]
in Python 2.5.2 (at least).

[terryjreedy]

2.5 is closed to fixes other than security fixes.
So this needs to be checked with the latest 2.6.

The implementation of new features does not always exactly follow the
PEP. So if there is also no warning in 2.6, the question remains whether
accident or intention. Perhaps it was decided that the warning was too
difficult or that it would make too much noise because some people were
doing this intentionally and would not want the warning, whereas few
people accidentally do the sort of thing your example does.

If this is true, as I sort of suspect, the PEP should be revised not to
mislead.

[terryjreedy]

Not an issue in 3.x.