Make digest_size a class variable

[rhettinger]

BPO	26700
Nosy	@rhettinger, @gpshead, @tiran

^{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 2017-05-06.22:50:43.158>
created_at = <Date 2016-04-06.07:59:10.344>
labels = ['extension-modules', 'type-feature']
title = 'Make digest_size a class variable'
updated_at = <Date 2017-05-06.22:50:43.156>
user = 'https://github.com/rhettinger'

bugs.python.org fields:

activity = <Date 2017-05-06.22:50:43.156>
actor = 'gregory.p.smith'
assignee = 'none'
closed = True
closed_date = <Date 2017-05-06.22:50:43.158>
closer = 'gregory.p.smith'
components = ['Extension Modules']
creation = <Date 2016-04-06.07:59:10.344>
creator = 'rhettinger'
dependencies = []
files = []
hgrepos = []
issue_num = 26700
keywords = []
message_count = 5.0
messages = ['262940', '266761', '266762', '275441', '293175']
nosy_count = 3.0
nosy_names = ['rhettinger', 'gregory.p.smith', 'christian.heimes']
pr_nums = []
priority = 'normal'
resolution = 'rejected'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue26700'
versions = ['Python 3.6']

[rhettinger]

It would be nicer if this worked:

     >>> hashlib.md5.digest_size
     64

[gpshead]

+cc Christian Heimes: do the upcoming standard hash functions amenable to a constant class attribute digest_size?

Are any of the expected future hashes variable length from a single class / constructor function?

[tiran]

Not all hash functions have a fixed digest size, e.g. blake2 is a variable length hash algorithm. https://pythonhosted.org/pyblake2/examples.html#using-different-digest-sizes How about a new class attribute default_digest_size?

[gpshead]

I agree that adding a .default_digest_size attribute to the constructors in the hashlib module would be good... the challenge is in actually doing it. Today they are not classes, they are built-in functions that come from one of two possible extension modules. Adding properties to a built-in function is a challenge (I don't think that is even possible in the C API, PyMethodDef doesn't support that).

We could hack around it, wrap them in classes, etc. But they have never been classes and there is little good reason for them to be a type.

[gpshead]

I'm rejecting this one as low value / infeasible.