python-peps/pep-0558.rst

PEP: 558
Title: Defined semantics for locals()
Author: Nick Coghlan <ncoghlan@gmail.com>
Status: Draft
Type: Standards Track
Content-Type: text/x-rst
Created: 2017-09-08
Python-Version: 3.7
Post-History: 2017-09-08


Abstract
========

The semantics of the ``locals()`` builtin have historically been underspecified
and hence implementation dependent.

This PEP proposes formally standardising on the behaviour of the CPython 3.6
reference implementation for most execution scopes, with some adjustments to the
behaviour at function scopes to make it more predictable and independent of the
presence or absence of tracing functions.


Rationale
=========

While the precise semantics of the ``locals()`` builtin are nominally undefined,
in practice, many Python programs depend on it behaving exactly as it behaves in
CPython (at least when no tracing functions are installed).

Other implementations such as PyPy are currently replicating that behaviour,
up to and including replication of local variable mutation bugs that
can arise when a trace hook is installed [1]_.


Proposal
========

The expected semantics of the ``locals()`` builtin change based on the current
execution scope. For this purpose, the defined scopes of execution are:

* module scope: top-level module code, as well as any other code executed using
  ``exec()`` or ``eval()`` with a single namespace
* class scope: code in the body of a ``class`` statement, as well as any other
  code executed using ``exec()`` or ``eval()`` with separate local and global
  namespaces
* function scope: code in the body of a ``def`` or ``async def`` statement


Module scope
------------

At module scope, as well as when using ``exec()`` or ``eval()`` with a
single namespace, ``locals()`` must return the same object as ``globals()``,
which must be the actual execution namespace (available as
``inspect.currentframe().f_locals`` in implementations that provide access
to frame objects).

Variable assignments during subsequent code execution in the same scope must
dynamically change the contents of the returned mapping, and changes to the
returned mapping must change the values bound to local variable names in the
execution environment.

This part of the proposal does not require any changes to the reference
implementation - it is standardisation of the current behaviour.


Class scope
-----------

At class scope, as well as when using ``exec()`` or ``eval()`` with separate
global and local namespaces, ``locals()`` must return the specified local
namespace (which may be supplied by the metaclass ``__prepare__`` method
in the case of classes). As for module scope, this must be a direct reference
to the actual execution namespace (available as
``inspect.currentframe().f_locals`` in implementations that provide access
to frame objects).

Variable assignments during subsequent code execution in the same scope must
change the contents of the returned mapping, and changes to the returned mapping
must change the values bound to local variable names in the
execution environment.

For classes, this mapping will *not* be used as the actual class namespace
underlying the defined class (the class creation process will copy the contents
to a fresh dictionary that is only accessible by going through the class
machinery).

This part of the proposal does not require any changes to the reference
implementation - it is standardisation of the current behaviour.


Function scope
--------------

At function scope, interpreter implementations are granted significant freedom
to optimise local variable access, and hence are NOT required to permit
arbitrary modification of local and nonlocal variable bindings through the
mapping returned from ``locals()``.

Instead, ``locals()`` is expected to return a mutable *snapshot* of the
function's local variables and any referenced nonlocal cells with the following
semantics:

* each call to ``locals()`` returns the *same* mapping object
* each call to ``locals()`` updates the mapping to the current
  state of the local variables and any referenced nonlocal cells
* changes to the returned mapping are *never* written back to the
  local variable bindings or the nonlocal cell references
* in implementations that provide access to frame objects, the return value
  from ``locals()`` is *not* a direct reference to
  ``inspect.currentframe().f_locals``

For interpreters that provide access to frame objects, the ``frame.f_locals``
attribute at function scope is expected to be a write-through proxy that
immediately updates the local variables and reference nonlocal cell bindings.
Additional entries may also be added to ``frame.f_locals`` and will be
accessible through both ``frame.f_locals`` and ``locals()`` from inside the
frame, but will not be accessible by name from within the function (as any
names which don't appear as local or nonlocal variables at compile time will
only be looked up in the module globals and process builtins, not in the
function locals).


Open Questions
==============

Allowing local variable binding mutation outside trace functions
----------------------------------------------------------------

This PEP allows local variable bindings to be mutated whenever code has access
to the frame object - it doesn't restrict that ability to trace functions the
way the status quo does.

It does this since it wants to allow trace functions to make changes, while
removing the current bulk copy from ``f_locals`` back to the frame state when
the trace function returns.

An alternative approach might be to *temporarily* replace ``f_locals`` with
a write-through proxy while the trace function is running, and then swap it
back to the result of ``locals()`` when the trace function returns.


Where is the new ``locals()`` result stored?
--------------------------------------------

If ``locals()`` is a new mapping distinct from the write-through proxy stored in
``frame.f_locals``, where will that mapping be stored?

A new lazily initialised frame attribute seems like a plausible answer, but that
raises new questions around how that attribute will be managed for module and
class scopes (set to the same thing as ``f_locals``? Set to ``NULL``/``None``?)

Alternatively, it could be stored in ``f_locals`` most of the time (as it is
today), and have the write-through proxy stored in a separate lazily
initialised attribute that gets swapped in as ``f_locals`` only when calling
trace functions.


What happens with the default args for ``eval()`` and ``exec()``?
-----------------------------------------------------------------

These are formally defined as inheriting ``globals()`` and ``locals()`` from
the calling scope by default.

There doesn't seem to be any reason for the PEP to change this.


Does mutating the ``f_locals`` proxy refresh the ``locals()`` mapping?
----------------------------------------------------------------------

This is probably needed in order to retain the current behaviour where writes
to ``frame.f_locals`` are immediately visible via references obtained via
``locals()``.


How much compatibility is enough compatibility?
-----------------------------------------------

As discussed below, the proposed design aims to keep almost all current code
working, *except* code that relies on being able to mutate function local
variable bindings and nonlocal cell references via the ``locals()`` builtin
when a trace hook is installed.

This is considered reasonable, as if a trace hook is installed, that indicates
the use of an interpreter implementation that provides access to frame objects,
and hence ``frame.f_locals`` can be used as a more portable and future-proof
alternative.

If some other existing behaviours are deemed optional (e.g. allowing
``locals()`` to return a fresh object each time), then that may allow for
some simplification of the update implementation.


Design Discussion
=================

Making ``locals()`` return a shared snapshot at function scope
--------------------------------------------------------------

The ``locals()`` builtin is a required part of the language, and in the
reference implementation it has historically returned a mutable mapping with
the following characteristics:

* each call to ``locals()`` returns the *same* mapping
* each call to ``locals()`` updates the mapping with the current
  state of the local variables and any referenced nonlocal cells
* changes to the returned mapping *usually* aren't written back to the
  local variable bindings or the nonlocal cell references, but write backs
  can be triggered by doing one of the following:

  * installing a Python level trace hook (write backs then happen whenever
    the trace hook is called)
  * running a function level wildcard import (requires bytecode injection in Py3)
  * running an ``exec`` statement in the function's scope (Py2 only, since
    ``exec`` became an ordinary builtin in Python 3)

The current proposal aims to retain the first two properties (to maintain
backwards compatibility with as much code as possible) while still
eliminating the ability to dynamically alter local and nonlocal variable
bindings through the mapping returned by ``locals()``.


Making ``frame.f_locals`` a write-through proxy at function scope
-----------------------------------------------------------------

While frame objects and related APIs are an explicitly optional feature of
Python implementations, there are nevertheless a lot of debuggers and other
introspection tools that expect them to behave in certain ways, including the
ability to update the bindings of local variables and nonlocal cell references,
as well as being able to store custom keys in the locals namespace for
arbitrary frames and retrieve those values later.

CPython currently supports this by copying the local variable bindings and
nonlocal cell references to ``frame.f_locals`` before calling a trace function,
and then copying them back after the function returns.

Unfortunately, as documented in [1]_, this approach leads to intrinsic race
conditions when a trace function writes to a closure variable via
``frame.f_locals``.

Switching to immediate updates of the frame state via ``frame.f_locals`` means
that the behaviour of trace functions should be more predictable, even in the
presence of multi-threaded access.


Historical semantics at function scope
--------------------------------------

The current semantics of mutating ``locals()`` and ``frame.f_locals`` in CPython
are rather quirky due to historical implementation details:

* actual execution uses the fast locals array for local variable bindings and
  cell references for nonlocal variables
* there's a ``PyFrame_FastToLocals`` operation that populates the frame's
  ``f_locals`` attribute based on the current state of the fast locals array
  and any referenced cells. This exists for three reasons:

  * allowing trace functions to read the state of local variables
  * allowing traceback processors to read the state of local variables
  * allowing locals() to read the state of local variables
* a direct reference to ``frame.f_locals`` is returned from ``locals()``, so if
  you hand out multiple concurrent references, then all those references will be
  to the exact same dictionary
* the two common calls to the reverse operation, ``PyFrame_LocalsToFast``, were
  removed in the migration to Python 3: ``exec`` is no longer a statement (and
  hence can no longer affect function local namespaces), and the compiler now
  disallows the use of ``from module import *`` operations at function scope
* however, two obscure calling paths remain: ``PyFrame_LocalsToFast`` is called
  as part of returning from a trace function (which allows debuggers to make
  changes to the local variable state), and you can also still inject the
  ``IMPORT_STAR`` opcode when creating a function directly from a code object
  rather than via the compiler

This proposal deliberately *doesn't* formalise these semantics as is, since they
only make sense in terms of the historical evolution of the language and the
reference implementation, rather than being deliberately designed.

Rejected Alternatives
=====================

Making ``locals()`` return the write-through proxy directly
-----------------------------------------------------------

A number of changes have been made to ``locals()`` over the years to
deliberately make it *harder* for arbitrary code to mutate function local
variables without those changes being visible to the compiler at compile time.

As such the desired default semantics for ``locals()`` are those currently
seen when a tracing function *isn't* installed, rather than the mutating
behaviour currently seen when a tracing hook is installed.


Implementation
==============

The reference implementation update is TBD - when available, it will be linked
from [2]_.

References
==========

.. [1] Broken local variable assignment given threads + trace hook + closure
   (https://bugs.python.org/issue30744)

.. [2] Clarify the required behaviour of ``locals()``
   (https://bugs.python.org/issue17960)

Copyright
=========

This document has been placed in the public domain.



..
   Local Variables:
   mode: indented-text
   indent-tabs-mode: nil
   sentence-end-double-space: t
   fill-column: 70
   coding: utf-8
   End: