Inconsistent 'file' vs 'stream' kwarg in pprint, other stdlibs

[pjenvey]

BPO	6226
Nosy	@pjenvey, @merwok, @bitdancer

^{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 2010-11-12.02:29:23.899>
created_at = <Date 2009-06-06.22:02:07.660>
labels = ['type-feature', 'library']
title = "Inconsistent 'file' vs 'stream' kwarg in pprint, other stdlibs"
updated_at = <Date 2010-11-12.02:29:23.898>
user = 'https://github.com/pjenvey'

bugs.python.org fields:

activity = <Date 2010-11-12.02:29:23.898>
actor = 'eric.araujo'
assignee = 'none'
closed = True
closed_date = <Date 2010-11-12.02:29:23.899>
closer = 'eric.araujo'
components = ['Library (Lib)']
creation = <Date 2009-06-06.22:02:07.660>
creator = 'pjenvey'
dependencies = []
files = []
hgrepos = []
issue_num = 6226
keywords = []
message_count = 4.0
messages = ['89020', '112300', '121002', '121009']
nosy_count = 3.0
nosy_names = ['pjenvey', 'eric.araujo', 'r.david.murray']
pr_nums = []
priority = 'normal'
resolution = 'wont fix'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue6226'
versions = ['Python 3.2']

[pjenvey]

It'd be nice to eventually standardize on the kwarg name used for basic
file-like args in the stdlib. print, warnings.showwarning and some
others take a file= argument whereas pprint, getpass.getpass take
stream=

print and pprint in particular should match -- though they do have a
different option set, when you're using the same options this
consistency would ease replacing:

print(obj, file=sys.stderr)
with
pprint(obj, stream=sys.stderr)

[BreamoreBoy]

@philip can you provide a patch so we can take this forward?

[merwok]

I’m afraid backward compatibility binds us here. For new code however, it would be nice to use one name consistently. My preference goes to “file”, because I really like the print function, but I suspect this could turn into bikeshedding quickly :) I think this is too little an issue to warrant a line in the style guide. Maybe just watch python-checkins and comment when someone adds a function using “stream”, “fp” or “fileobj”. I suggest closing this report as wontfix.

You may want to open a new report about the matter of pprint and print having incompatible signatures. How about proposing a new function, say pprint.print, which could be a drop-in replacement for builtins.print?

[bitdancer]

Yes, simply saying "make it consistent" is a won't fix, because of backward compatibility issues. Specific proposals for specific functions in a way that deals with the backward compatibility issues should be opened as new issues.