Victor Stinner [Thu, 3 Jun 2021 16:42:59 +0000 (18:42 +0200)]
bpo-39573: Py_TYPE becomes a static inline function (GH-26493)
Convert the Py_TYPE() and Py_SIZE() macros to static inline
functions. The Py_SET_TYPE() and Py_SET_SIZE() functions must now be
used to set an object type and size.
Eric Snow [Thu, 3 Jun 2021 16:28:27 +0000 (10:28 -0600)]
bpo-43693: Add new internal code objects fields: co_fastlocalnames and co_fastlocalkinds. (gh-26388)
A number of places in the code base (notably ceval.c and frameobject.c) rely on mapping variable names to indices in the frame "locals plus" array (AKA fast locals), and thus opargs. Currently the compiler indirectly encodes that information on the code object as the tuples co_varnames, co_cellvars, and co_freevars. At runtime the dependent code must calculate the proper mapping from those, which isn't ideal and impacts performance-sensitive sections. This is something we can easily address in the compiler instead.
This change addresses the situation by replacing internal use of co_varnames, etc. with a single combined tuple of names in locals-plus order, along with a minimal array mapping each to its kind (local vs. cell vs. free). These two new PyCodeObject fields, co_fastlocalnames and co_fastllocalkinds, are not exposed to Python code for now, but co_varnames, etc. are still available with the same values as before (though computed lazily).
Aside from the (mild) performance impact, there are a number of other benefits:
* there's now a clear, direct relationship between locals-plus and variables
* code that relies on the locals-plus-to-name mapping is simpler
* marshaled code objects are smaller and serialize/de-serialize faster
Also note that we can take this approach further by expanding the possible values in co_fastlocalkinds to include specific argument types (e.g. positional-only, kwargs). Doing so would allow further speed-ups in _PyEval_MakeFrameVector(), which is where args get unpacked into the locals-plus array. It would also allow us to shrink marshaled code objects even further.
The PyType_Ready() function now raises an error if a type is defined
with the Py_TPFLAGS_HAVE_GC flag set but has no traverse function
(PyTypeObject.tp_traverse).
Allocate and track statement objects in pysqlite_statement_create.
By allocating and tracking creation of statement object in
pysqlite_statement_create(), the caller does not need to worry about GC
syncronization, and eliminates the possibility of getting a badly
created object. All related fault handling is moved to
pysqlite_statement_create().
Co-authored-by: Victor Stinner <vstinner@python.org>
Victor Stinner [Mon, 31 May 2021 11:10:31 +0000 (13:10 +0200)]
bpo-44263: Fix _decimal and _testcapi GC protocol (GH-26464)
* _testcapi.heapgctype: implement a traverse function since the type
is defined with Py_TPFLAGS_HAVE_GC.
* _decimal: PyDecSignalDictMixin_Type is no longer defined with
Py_TPFLAGS_HAVE_GC since it has no traverse function.
Trivial change in fractions module docs: real -> rational numbers (GH-25009)
Obviously, the former was a little misleading: not only rationals
may be considered as "infinite-precision, real numbers" - but, for
example, any real finite extension of the rationals.
Terry Jan Reedy [Sat, 29 May 2021 07:19:50 +0000 (03:19 -0400)]
bpo-44254: On Mac, remove disfunctional colors from turtledemo buttons (GH-26448)
On macOS, tk defers to system setting for button background when in normal state.
Give turtledemo button text a color that works on either light or dark background.
Jason R. Coombs [Thu, 27 May 2021 23:36:06 +0000 (19:36 -0400)]
bpo-44246: Update What's New for importlib.metadata. (#26408)
* bpo-44246: Update What's New for importlib.metadata.
Bump version of importlib_metadata included.
Add note about compatibility notice and fix link to entry_points documentation.
Add note about removal of access by index on Distribution.entry_points.
* Fix syntax mistake in issue reference.
Co-authored-by: Ken Jin <28750310+Fidget-Spinner@users.noreply.github.com>
* Fix broken reference in entry-points.
Co-authored-by: Ken Jin <28750310+Fidget-Spinner@users.noreply.github.com>
Fix a regression in type() when a metaclass raises an exception. The
C function type_new() must properly report the exception when a
metaclass constructor raises an exception and the winner class is not
the metaclass.
Petr Viktorin [Tue, 25 May 2021 11:42:03 +0000 (13:42 +0200)]
bpo-43795: Remove Py_FrozenMain from the Limited API & Stable ABI (GH-26241)
Py_FrozenMain was added to the Limited C API in [bpo-42591]() (3.10.0a4);
but to fix that issue it would be enough to add it to the regular C API.
The function is undocumented, tests were added very recently ([bpo-44131]()),
and most importantly, it is not present in all builds of Python, as
the linker sometimes omits it as unused.
It should be added back when these issues are fixed.
Note that this does not affect Python's regular C API.
Ken Jin [Mon, 24 May 2021 23:51:08 +0000 (07:51 +0800)]
bpo-38908: [docs] Add changes to 3.10 whatsnew and fix some minor inaccuracies in news (GH-26096)
The fix only applies to ``isinstance``. ``issubclass`` isn't affected (because it was always working to begin with). So I also fixed the news to reflect that.
Petr Viktorin [Mon, 24 May 2021 22:48:44 +0000 (00:48 +0200)]
bpo-41282: Fix broken `make install` (GH-26329)
A previous commit broke a check in sysconfig when building cpython itself.
This caused builds of the standard library modules to search a wrong
location (the installed location rather than the source directory) for
header files with the net effect that a ``make install``
incorrectly caused all extension modules to be rebuilt again and
with incorrect include file paths.
When building Python, we need two distinct "include" directories:
- source .h files
- install target for .h files
Note that this doesn't matter except when building Python from source.
Historically:
- source .h files were in the distutils scheme under 'include'
- the install directory was in the distutils.command.install scheme
under 'headers'
GH-24549 merged these; sysconfig is now the single source of truth and
distutils is derived from it.
This commit introduces a "secret" scheme path, 'headers', which contains
the install target. It is only present when building Python.
The distutils code uses it if present, and falls back to 'include'.
Pablo Galindo [Sat, 22 May 2021 22:05:00 +0000 (23:05 +0100)]
bpo-44201: Avoid side effects of "invalid_*" rules in the REPL (GH-26298)
When the parser does a second pass to check for errors, these rules can
have some small side-effects as they may advance the parser more than
the point reached in the first pass. This can cause the tokenizer to ask
for extra tokens in interactive mode causing the tokenizer to show the
prompt instead of failing instantly.
To avoid this, add a new mode to the tokenizer that is activated in the
second pass and deactivates asking for new tokens when the interactive
line is finished. As the parsing should have reached the last line in
the first pass, the second pass should not need to ask for more tokens.