HTTPConnection.requests has a mutable as default value for headers

[matrixise]

BPO	27728
Nosy	@bitdancer, @vadmium, @matrixise
Files	issue27728.diff

^{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 2017-05-24.12:42:53.788>
created_at = <Date 2016-08-10.17:58:34.291>
labels = ['invalid']
title = 'HTTPConnection.requests has a mutable as default value for headers'
updated_at = <Date 2017-05-24.12:42:53.787>
user = 'https://github.com/matrixise'

bugs.python.org fields:

activity = <Date 2017-05-24.12:42:53.787>
actor = 'matrixise'
assignee = 'none'
closed = True
closed_date = <Date 2017-05-24.12:42:53.788>
closer = 'matrixise'
components = []
creation = <Date 2016-08-10.17:58:34.291>
creator = 'matrixise'
dependencies = []
files = ['44070']
hgrepos = []
issue_num = 27728
keywords = ['patch']
message_count = 7.0
messages = ['272352', '272353', '272356', '272359', '272364', '272380', '272389']
nosy_count = 4.0
nosy_names = ['r.david.murray', 'SilentGhost', 'martin.panter', 'matrixise']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue27728'
versions = ['Python 3.6']

[matrixise]

http.client.HTTPConnection.request, the argument 'headers' has a default value, and this one is a mutable (dict).

[matrixise]

here is my patch for that.

[SilentGhost]

Why is that an issue?

[bitdancer]

Well, it's a problem in general because if the called routine modifies the default dict...the default is modified. I'm guessing it is never modified, but it is indeed better future proofing of the code to not use a mutable default in a routine that passes the variable to other functions, as this one does.

[matrixise]

Thanks R.David.Murray,

@SilentGhost in fact, in this code, it's not an issue but we use a bad practice and not a good practice with a mutable value.

[vadmium]

If the code does happen to mutate the dictionary, there would still be a bug with modifying any dictionary passed in by the caller. But I suspect there is no bug. So I am not sure the change is worth it. If we do go ahead, the documentation needs updating to say None is now allowed.

[bitdancer]

Yes, this is one of those cases where the benefit of the change is probably outweighed by the cost.