For Loop temporary variable scope should be local to For loop

[debgit]

BPO	44332
Nosy	@stevendaprano, @debgit

^{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 2021-06-07.06:58:19.496>
created_at = <Date 2021-06-07.04:56:39.464>
labels = ['invalid', 'type-bug', 'library', '3.9']
title = 'For Loop temporary variable scope should be local to For loop'
updated_at = <Date 2021-06-07.06:58:19.496>
user = 'https://github.com/debgit'

bugs.python.org fields:

activity = <Date 2021-06-07.06:58:19.496>
actor = 'mark.dickinson'
assignee = 'none'
closed = True
closed_date = <Date 2021-06-07.06:58:19.496>
closer = 'mark.dickinson'
components = ['Library (Lib)']
creation = <Date 2021-06-07.04:56:39.464>
creator = 'deb_ctc'
dependencies = []
files = []
hgrepos = []
issue_num = 44332
keywords = []
message_count = 3.0
messages = ['395246', '395247', '395248']
nosy_count = 2.0
nosy_names = ['steven.daprano', 'deb_ctc']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue44332'
versions = ['Python 3.9']

[debgit]

numbers = [1, 2, 3, 4, 5, 6, 7, 8]
for x in numbers:
    print(x)
print(x)

In the above code, print(x) statement should give error. Because x scope should be local to for loop only.
99% cases, developers will not use the temporary variable x outside of the for loop.
So x will keep on consuming memory always. So it is a bad usage of memory.
Ideally x memory should be free once for loop execution is completed.

[stevendaprano]

This is not a bug, it is intentional design and has been since Python 1.

You are incorrect about x consuming memory "always". If the value bound to x is in use elsewhere, deleting x will save no memory. If the value is not in use elsewhere, it will be garbage collected as soon as x goes out of scope, which will be at the end of the function.

Shifting to block-scope for for-loops has been discussed before, it is not a popular idea and I expect that most people will oppose it. You can search the Python-Ideas mailing list if you want to find out more. If you still feel strongly that this is an enhancement, as it is a major change in language behaviour it would require a PEP to be written.

https://www.python.org/dev/peps/pep-0001/

Even if the PEP was accepted, the earliest it could go into the language would be Python 3.11 or better, and that would likely require a special __future__ import.

[stevendaprano]

By the way, loop variables are not considered to be "temporary" in Python. They are no more temporary than any other local variable -- they *are* local variables with exactly the same scope and lifetime as all other local variables.

[victorel-petrovich]

Anyone think it would be good to have this option:

for local i in [1]:
   m=4  # assigns to existing m or creates new if none
   local j = 9 # makes j only valid for current suite 
   
print(m) # 4
print(i) # error
print(j) # error