[Import-SIG] Proto-PEP: Redesigning extension module loading
encukou at gmail.com
Mon Mar 2 15:21:19 CET 2015
>>>> We should expose some kind of API in importlib.util (or a better place?)
>>>> can be used to check that a module works with reloading and
>>> What would such an API actually check to verify that a module could be
>> Obviously we can't check for static state or object leakage between
>> By using the new API, you promise that the extension does support
>> reloading and subinterpreters. This will be prominently stated in the
>> docs, and checked by this function.
>> For the old API, PyModule_Create with m_size>=0 can be used to support
>> subinterpreters. But I don't think the language in the docs is strong
>> enough to say that m_size>=0 is a promise of such support.
> Ah, I wasn't clear in terms of "check" or "test" when I mentioned this
> - I was literally referring to something that could be run in test
> suites to try these things and see if they worked or not, rather than
> to a runtime "can I reload this safely?" check. "Try it and see" is
> likely to be a better approach to take there.
Hm, how would such a test work?
A function that takes a piece of code (like timeit does), runs it in a
new subinterpreter, and check for leaks? Or runs it in a new process
and verifies no objects remain after PyFinalize?
That seems way out of scope here.
Here is a new draft.
I have removed the "Create-only" option, which simplified the PEP a bit.
I've added PyCapsule helper functions. These ended up taking quite a
It would be possible to derive the capsule name just from module.__name__
and the attribute name, following the PyCapsule_Import convention,
but I think specifying it explicitly is necessary to get the proper
I ended up requiring the module name, and constructing the capsule name
from that and the attribute. So I got:
const char *module_name,
const char *attribute_name,
const char *module_name,
const char *attribute_name)
The first one would usually be used once per module, and the second one
begs for an extension-specific macro to cast the result to a usable type,
so expected usage is just SPAM_GET_DATA(m).
I think this draft is fine now so I'll start working on the implementation:
Title: Redesigning extension module loading
Author: Petr Viktorin <encukou at gmail.com>, Stefan Behnel <stefan_ml
at behnel.de>, Nick Coghlan <ncoghlan at gmail.com>
Type: Standards Track
Post-History: 23-Aug-2013, 20-Feb-2015
This PEP proposes a redesign of the way in which extension modules interact
with the import machinery. This was last revised for Python 3.0 in PEP
3121, but did not solve all problems at the time. The goal is to solve them
by bringing extension modules closer to the way Python modules behave;
specifically to hook into the ModuleSpec-based loading mechanism
introduced in PEP 451.
Extensions that do not require custom memory layout for their module objects
may be executed in arbitrary pre-defined namespaces, paving the way for
extension modules being runnable with Python's ``-m`` switch.
Other extensions can use custom types for their module implementation.
Module types are no longer restricted to types.ModuleType.
This proposal makes it easy to support properties at the module
level and to safely store arbitrary global state in the module that is
covered by normal garbage collection and supports reloading and
Extension authors are encouraged to take these issues into account
when using the new API.
Python modules and extension modules are not being set up in the same way.
For Python modules, the module is created and set up first, then the module
code is being executed (PEP 302).
A ModuleSpec object (PEP 451) is used to hold information about the module,
and passed to the relevant hooks.
For extensions, i.e. shared libraries, the module
init function is executed straight away and does both the creation and
initialisation. The initialisation function is not passed ModuleSpec
information about the loaded module, such as the __file__ or fully-qualified
name. This hinders relative imports and resource loading.
This is specifically a problem for Cython generated modules, for which it's
not uncommon that the module init code has the same level of complexity as
that of any 'regular' Python module. Also, the lack of __file__ and __name__
information hinders the compilation of __init__.py modules, i.e. packages,
especially when relative imports are being used at module init time.
The other disadvantage of the discrepancy is that existing Python programmers
learning C cannot effectively map concepts between the two domains.
As long as extension modules are fundamentally different from pure Python ones
in the way they're initialised, they are harder for people to pick up without
relying on something like cffi, SWIG or Cython to handle the actual extension
Currently, extension modules are also not added to sys.modules until they are
fully initialized, which means that a (potentially transitive)
re-import of the module will really try to reimport it and thus run into an
infinite loop when it executes the module init function again.
Without the fully qualified module name, it is not trivial to correctly add
the module to sys.modules either.
Furthermore, the majority of currently existing extension modules has
problems with sub-interpreter support and/or reloading, and, while it is
possible with the current infrastructure to support these
features, it is neither easy nor efficient.
Addressing these issues was the goal of PEP 3121, but many extensions,
including some in the standard library, took the least-effort approach
to porting to Python 3, leaving these issues unresolved.
This PEP keeps the backwards-compatible behavior, which should reduce pressure
and give extension authors adequate time to consider these issues when porting.
The current process
Currently, extension modules export an initialisation function named
"PyInit_modulename", named after the file name of the shared library. This
function is executed by the import machinery and must return either NULL in
the case of an exception, or a fully initialised module object. The
function receives no arguments, so it has no way of knowing about its
During its execution, the module init function creates a module object
based on a PyModuleDef struct. It then continues to initialise it by adding
attributes to the module dict, creating types, etc.
In the back, the shared library loader keeps a note of the fully qualified
module name of the last module that it loaded, and when a module gets
created that has a matching name, this global variable is used to determine
the fully qualified name of the module object. This is not entirely safe as it
relies on the module init function creating its own module object first,
but this assumption usually holds in practice.
The current extension module initialisation will be deprecated in favour of
a new initialisation scheme. Since the current scheme will continue to be
available, existing code will continue to work unchanged, including binary
Extension modules that support the new initialisation scheme must export
the public symbol "PyModuleExec_modulename", and optionally
"PyModuleCreate_modulename", where "modulename" is the
name of the module. This mimics the previous naming convention for
the "PyInit_modulename" function.
If defined, these symbols must resolve to C functions with the following
int (*PyModuleExecFunction)(PyObject* module)
PyObject* (*PyModuleCreateFunction)(PyObject* module_spec)
The PyModuleExec function
The PyModuleExec function is used to implement "loader.exec_module"
defined in PEP 451.
It function will be called to initialize a module. (Usually, this amounts to
setting the module's initial attributes.)
This happens in two situations: when the module is first initialized for
a given (sub-)interpreter, and possibly later when the module is reloaded.
When PyModuleExec is called, the module has already been added to
sys.modules, and import-related attributes specified in
PEP 451 [#pep-0451-attributes]_) have been set on the module.
The "module" argument receives the module object to initialize.
If PyModuleCreate is defined, "module" will generally be the the object
returned by it.
It is possible for a custom loader to pass any object to
PyModuleExec, so this function should check and fail with TypeError
if the module's type is unsupported.
Any other assumptions should also be checked.
If PyModuleCreate is not defined, PyModuleExec is expected to operate
on any Python object for which attributes can be added by PyObject_GetAttr*
and retrieved by PyObject_SetAttr*.
This allows loading an extension into a pre-created module, making it possible
to run it as __main__ in the future, participate in certain lazy-loading
schemes [#lazy_import_concerns]_, or enable other creative uses.
If PyModuleExec replaces the module's entry in sys.modules,
the new object will be used and returned by importlib machinery.
(This mirrors the behavior of Python modules. Note that for extensions,
implementing PyModuleCreate is usually a better solution for the use cases
The function must return ``0`` on success, or, on error, set an exception and
The PyModuleCreate function
The optional PyModuleCreate function is used to implement
"loader.create_module" defined in PEP 451.
By exporting it, an extension module indicates that it uses a custom
This prevents loading the extension in a pre-created module,
but gives greater flexibility in allowing a custom C-level layout
of the module object.
Most extensions will not need to implement this function.
The "module_spec" argument receives a "ModuleSpec" instance, as defined in
When called, this function must create and return a module object,
or set an exception and return NULL.
There is no requirement for the returned object to be an instance of
types.ModuleType. Any type can be used, as long as it supports setting and
getting attributes, including at least the import-related attributes
specified in PEP 451 [#pep-0451-attributes]_.
This follows the current support for allowing arbitrary objects in sys.modules
and makes it easier for extension modules to define a type that exactly matches
their needs for holding module state.
Note that when this function is called, the module's entry in sys.modules
is not populated yet. Attempting to import the same module again
(possibly transitively), may lead to an infinite loop.
Extension authors are advised to keep PyModuleCreate minimal, an in particular
to not call user code from it.
If PyModuleCreate is not defined, the default loader will construct
a module object as if with PyModule_New.
Initialization helper functions
For two initialization tasks previously done by PyModule_Create,
two functions are introduced::
int PyModule_SetDocString(PyObject *m, const char *doc)
int PyModule_AddFunctions(PyObject *m, PyMethodDef *functions)
These set the module docstring, and add the module functions, respectively.
Both will work on any Python object that supports setting attributes.
They return ``0`` on success, and on failure, they set the exception
and return ``-1``.
PyCapsule convenience functions
Instead of custom module objects, PyCapsule will become the preferred
mechanism for storing per-module C data.
Two new convenience functions will be added to help with this.
const char *module_name,
const char *attribute_name,
Add a new PyCapsule to *module* as *attribute_name*.
The capsule name is formed by joining *module_name* and *attribute_name*
by a dot.
This convenience function can be used from a module initialization function
instead of separate calls to PyCapsule_New and PyModule_AddObject.
Returns a borrowed reference to the new capsule,
or NULL (with exception set) on failure.
const char *module_name,
const char *attribute_name)
Returns the pointer stored in *module* as *attribute_name*, or NULL
(with an exception set) on failure. The capsule name is formed by joining
*module_name* and *attribute_name* by a dot.
This convenience function can be used instead of separate calls to
PyObject_GetAttr and PyCapsule_GetPointer.
Extension authors are encouraged to define a macro to
call PyModule_GetCapsulePointer and cast the result to an appropriate type.
Generalizing PyModule_* functions
The following functions and macros will be modified to work on any object
that supports attribute access:
The PyModule_GetDict function will continue to only work on true module
objects. This means that it should not be used on extension modules that only
If PyModuleExec is not defined, the import machinery will try to initialize
the module using the PyModuleInit hook, as described in PEP 3121.
If PyModuleExec is defined, PyModuleInit will be ignored.
Modules requiring compatibility with previous versions of CPython may implement
PyModuleInit in addition to the new hook.
Subinterpreters and Interpreter Reloading
Extensions using the new initialization scheme are expected to support
subinterpreters and multiple Py_Initialize/Py_Finalize cycles correctly.
The mechanism is designed to make this easy, but care is still required
on the part of the extension author.
No user-defined functions, methods, or instances may leak to different
To achieve this, all module-level state should be kept in either the module
dict, or in the module object.
A simple rule of thumb is: Do not define any static data, except built-in types
with no mutable or user-settable class attributes.
Reloading an extension module will re-execute its PyModuleInit function.
Similar caveats apply to reloading an extension module as to reloading
a Python module. Notably, attributes or any other state of the module
are not reset before reloading.
Additionally, due to limitations in shared library loading (both dlopen on
POSIX and LoadModuleEx on Windows), it is not generally possible to load
a modified library after it has changed on disk.
Therefore, reloading extension modules is of limited use.
Multiple modules in one library
To support multiple Python modules in one shared library, the library
must export appropriate PyModuleExec_<name> or PyModuleCreate_<name> hooks
for each exported module.
The modules are loaded using a ModuleSpec with origin set to the name of the
library file, and name set to the module name.
Note that this mechanism can currently only be used to *load* such modules,
not to *find* them.
XXX: This is an existing issue; either fix it/wait for a fix or provide
an example of how to load such modules.
XXX - not started
We should expose some kind of API in importlib.util (or a better place?) that
can be used to check that a module works with reloading and subinterpreters.
The runpy module will need to be modified to take advantage of PEP 451
and this PEP. This is out of scope for this PEP.
Stefan Behnel's initial proto-PEP [#stefans_protopep]_
had a "PyInit_modulename" hook that would create a module class,
whose ``__init__`` would be then called to create the module.
This proposal did not correspond to the (then nonexistent) PEP 451,
where module creation and initialization is broken into distinct steps.
It also did not support loading an extension into pre-existing module objects.
Nick Coghlan proposed the Create annd Exec hooks, and wrote a prototype
At this time PEP 451 was still not implemented, so the prototype
does not use ModuleSpec.
This document has been placed in the public domain.
More information about the Import-SIG