Document the rules for "public names"

[abalkin]

BPO	10434
Nosy	@freddrake, @warsaw, @ncoghlan, @abalkin, @ezio-melotti, @voidspace, @MaxwellDupre

^{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 = None
created_at = <Date 2010-11-16.14:55:37.412>
labels = ['easy', 'type-feature', 'docs']
title = 'Document the rules for "public names"'
updated_at = <Date 2022-01-21.17:33:05.594>
user = 'https://github.com/abalkin'

bugs.python.org fields:

activity = <Date 2022-01-21.17:33:05.594>
actor = 'mikecmcleod'
assignee = 'docs@python'
closed = False
closed_date = None
closer = None
components = ['Documentation']
creation = <Date 2010-11-16.14:55:37.412>
creator = 'belopolsky'
dependencies = []
files = []
hgrepos = []
issue_num = 10434
keywords = ['easy']
message_count = 5.0
messages = ['121297', '121299', '121304', '194067', '411158']
nosy_count = 9.0
nosy_names = ['fdrake', 'barry', 'ncoghlan', 'belopolsky', 'ron_adam', 'ezio.melotti', 'michael.foord', 'docs@python', 'mikecmcleod']
pr_nums = []
priority = 'normal'
resolution = None
stage = 'needs patch'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue10434'
versions = ['Python 3.2', 'Python 3.3', 'Python 3.4']

[abalkin]

As discussed in "Breaking undocumented API" thread [1] on python-dev, a definition of "public names" is buried deep in the language reference manual:

"""
The public names defined by a module are determined by checking the
module’s namespace for a variable named __all__; if defined, it must
be a sequence of strings which are names defined or imported by that
module. The names given in __all__ are all considered public and are
required to exist. If __all__ is not defined, the set of public names
includes all names found in the module’s namespace which do not begin
with an underscore character ('_'). __all__ should contain the entire
public API. It is intended to avoid accidentally exporting items that
are not part of the API (such as library modules which were imported
and used within the module).
""" [2]

It has been argued that this is not the authoritative definition and alternatives have been suggested such as "any name that does not begin with an underscore except imported modules."
mportant for the users and developers of cpython and other python implementations to know what names they can rely upon to stay defined between releases, the rules for "public names" should be documented.

I agree that the library manual is a more appropriate place for such documentation. The definition should include the naming conventions and the set of promises that Python makes about public name availability in the future releases.

[1] http://mail.python.org/pipermail/python-dev/2010-November/105490.html
[2] http://docs.python.org/reference/simple_stmts.html

[abalkin]

Michael Foord suggested adding the following to developer documentation such as PEP-8. [1] I am not sure PEP-8 is the right place for it. In my opinion, PEP-8 is mostly about stylistic choices that don't have a major impact on the users. In other words, unless you are developing python itself, you are unlikely to consult PEP-8. The API stability rules, however affect every user who programmed in python long enough to see a major release. I don't have objections to Michael's definitions below, but if included in the library manual, they should be reworded to address the user rather than the developer of the API.

"""
How about making this explicit (either PEP-8 or our developer docs):

If a module or package defines __all__ that authoritatively defines the
public interface. Modules with __all__ SHOULD still respect the naming
conventions (leading underscore for private members) to avoid confusing
users. Modules SHOULD NOT export private members in __all__.

Names imported into a module a never considered part of its public API
unless documented to be so or included in __all__.

Methods / functions / classes and module attributes whose names begin
with a leading underscore are private.

If a class name begins with a leading underscore none of its members are
public, whether or not they begin with a leading underscore.

If a module name in a package begins with a leading underscore none of
its members are public, whether or not they begin with a leading underscore.

If a module or package doesn't define __all__ then all names that don't
start with a leading underscore are public.

All public members MUST be documented. Public functions, methods and
classes SHOULD have docstrings. Private members may have docstrings.
""" [1]

[1] http://mail.python.org/pipermail/python-dev/2010-November/105476.html

[ronadam]

You may also want to update help topics.

help("PRIVATENAMES").

Identifiers (Names)

---

An identifier occurring as an atom is a name. See section
*Identifiers and keywords* for lexical definition and section *Naming
and binding* for documentation of naming and binding.

When the name is bound to an object, evaluation of the atom yields
that object. When a name is not bound, an attempt to evaluate it
raises a NameError exception.

Private name mangling: When an identifier that textually occurs in
a class definition begins with two or more underscore characters and
does not end in two or more underscores, it is considered a private
name of that class. Private names are transformed to a longer form
before code is generated for them. The transformation inserts the
class name in front of the name, with leading underscores removed, and
a single underscore inserted in front of the class name. For example,
the identifier __spam occurring in a class named Ham will be
transformed to _Ham__spam. This transformation is independent of
the syntactical context in which the identifier is used. If the
transformed name is extremely long (longer than 255 characters),
implementation defined truncation may happen. If the class name
consists only of underscores, no transformation is done.

Other topics that may be of interest.

IDENTIFIERS
NAMESPACES
PACKAGES
PRIVATENAMES
SPECIALATTRIBUTES
SPECIALIDENTIFIERS
SPECIALMETHODS

[ncoghlan]

PEP-8 now covers the developer side of things: http://www.python.org/dev/peps/pep-0008/#public-and-internal-interfaces

A user facing counterpart describing our backwards compatibility policy is still desirable. Updating PEP-5 wouldn't go astray, since that's *supposed* to serve that purpose :)

[MaxwellDupre]

I would like to help on this issue. Is there anyone available to push a PR through? If I make the changes.