bpo-40955: Fix memory leak in subprocess module (GH-20825)
```
Direct leak of 8 byte(s) in 1 object(s) allocated from:
GH-0 0x7f008bf19667 in __interceptor_malloc (/lib64/libasan.so.6+0xb0667)
GH-1 0x7f007a0bee4a in subprocess_fork_exec /home/heimes/dev/python/cpython/Modules/_posixsubprocess.c:774
GH-2 0xe0305b in cfunction_call Objects/methodobject.c:546
```
bpo-40847: Consider a line with only a LINECONT a blank line (GH-20769)
A line with only a line continuation character should be considered
a blank line at tokenizer level so that only a single NEWLINE token
gets emitted. The old parser was working around the issue, but the
new parser threw a `SyntaxError` for valid input. For example,
an empty line following a line continuation character was interpreted
as a `SyntaxError`.
bpo-40895: Update weakref documentation to remove old warnings (GH-20687)
The doccumentation at https://docs.python.org/3.10/library/weakref.html cautions that the `WeakKeyDictionary` and `WeakValueDictionary` are susceptible to the problem of dictionary mutation during iteration.
These notes present the user with a problem that has no easy solution.
I dug into the implementation and found that fortunately, Antoine Pitrou already addressed this challenge (10 years ago!) by introducing an `_IterationGuard` context manager to the implementation, which delays mutation while an iteration is in progress.
I asked for confirmation and @pitrou agreed that these notes could be removed:
https://github.com/python/cpython/commit/c1baa601e2b558deb690edfdf334fceee3b03327GH-commitcomment-39514438
(cherry picked from commit 1642c0ef750f96664a98cadb09301d492098d2fb)
Co-authored-by: Daniel Fortunov <asqui@users.noreply.github.com>
Victor Stinner [Wed, 10 Jun 2020 17:24:56 +0000 (19:24 +0200)]
bpo-40826: Fix test_repl.test_close_stdin() on Windows (GH-20779) (GH-20785)
test_repl.test_close_stdin() now calls
support.suppress_msvcrt_asserts() to fix the test on Windows.
* Move suppress_msvcrt_asserts() from test.libregrtest.setup to
test.support. Make its verbose parameter optional: verbose=False by
default.
* SuppressCrashReport now uses SetErrorMode() of the msvcrt module,
rather than using ctypes.
* Remove also an unused variable (deadline) in wait_process().
Victor Stinner [Mon, 8 Jun 2020 16:44:50 +0000 (18:44 +0200)]
bpo-40910: Export Py_GetArgcArgv() function (GH-20721) (GH-20723)
Export explicitly the Py_GetArgcArgv() function to the C API and
document the function. Previously, it was exported implicitly which
no longer works since Python is built with -fvisibility=hidden.
* Add PyConfig._orig_argv member.
* Py_InitializeFromConfig() no longer calls _PyConfig_Write() twice.
* PyConfig_Read() no longer initializes Py_GetArgcArgv(): it is now
_PyConfig_Write() responsibility.
* _PyConfig_Write() result type becomes PyStatus instead of void.
* Write an unit test on Py_GetArgcArgv().
bpo-33187: Document 3.9 changes to xml.etree.ElementInclude.include (GH-20438)
Looks like the merging of bpo-33187 and bpo-20928 was racy, resulting in
this change going undocumented.
(cherry picked from commit 301f0d4ff9b6bd60599eea0612904f65a92e6dd9)
allow macOS installer builds to package pre-built html docs (GH-20715)
build-installer now looks in its directory of source tarballs
for a suitable html tarball of the same version. If so, it
will unpack and use it rather than rebuilding the html format
documentation set from the source repo. This is intended as
a speedup for test builds of the installer. Files names must
be in the same format as produced by the docs build for download,
for example, `python-3.9.0b1-docs-html.tar.bz2`.
(cherry picked from commit 63fc55b2eab0331465605a49bfd28a1bcb997f92)
Before
```
content.txt is 42 bytes in size and isa regular file.
folder is 420 bytes in size and isa directory.
magic is 4200 bytes in size and issomething else.
```
After:
```
content.txt is 42 bytes in size and is a regular file.
folder is 420 bytes in size and is a directory.
magic is 4200 bytes in size and is something else.
```
bpo-40767: Allow pure Wayland to get default XDG web browser (GH-20382)
Would be nice to backport to python 3.7+. I don't think it's worth the hassle to backport this all the way down to 3.10. But I'll let the maintainers decide.
This is hard to test because the test setup already includes this [environment variable](https://github.com/python/cpython/blob/master/Lib/test/pythoninfo.pyGH-L292)
Let me know if something doesn't match the PR guidelines. This is my first PR in the python source code.
(cherry picked from commit c822efeda9a0afe87cf3429724732fc8e19a01fb)
Co-authored-by: Jeremy Attali <jeremy.attali@gmail.com>
The reset_peak function sets the peak memory size to the current size,
representing a resetting of that metric. This allows for recording the
peak of specific sections of code, ignoring other code that may have
had a higher peak (since the most recent `tracemalloc.start()` or
tracemalloc.clear_traces()` call).
Make sure that keyword arguments are merged into the arguments dictionary when dict unpacking and keyword arguments are interleaved. (GH-20553) (GH-20569)
Fix :mod:`ssl`` code to be compatible with OpenSSL 1.1.x builds that use
``no-deprecated`` and ``--api=1.1.0``.
Note: Tests assume full OpenSSL API and fail with limited API.
Signed-off-by: Christian Heimes <christian@python.org> Co-authored-by: Mark Wright <gienah@gentoo.org>
(cherry picked from commit a871f692b4a2e6c7d45579693e787edc0af1a02c)
Co-authored-by: Christian Heimes <christian@python.org>
bpo-17005: Move topological sort functionality to its own module (GH-20558)
The topological sort functionality that was introduced initially in the
functools module has been moved to a new graphlib module to
better accommodate the new tools and keep the original scope of the
functools module.
(cherry picked from commit 2f172d8f1525defe9bba4d49e967fdfc69151731)
Co-authored-by: Pablo Galindo <Pablogsal@gmail.com>
Fix asyncio.to_thread() documented return type (GH-20547)
When I wrote the documentation for `asyncio.to_thread()`, I mistakenly assumed that `return await loop.run_in_executor(...)` within an async def function would return a Future. In reality, it returns a coroutine.
This likely won't affect typical usage of `asyncio.to_thread()`, but it's important for the documentation to be correct here. In general, we also tend to avoid returning futures from high-level APIs in asyncio.
(cherry picked from commit 2b201369b435a4266bda5b895e3b615dbe28ea6e)
bpo-39885: Make IDLE context menu cut and copy work again (GH-18951)
Leave selection when right click within. This exception to clearing selections when right-clicking was omitted from the previous commit, 4ca060d. I did not realize that this completely disabled the context menu entries, and I should have merged a minimal fix immediately. An automated test should follow.
(cherry picked from commit 97e4e0f53d6690db6b942678489716a30925b8af)
Co-authored-by: Terry Jan Reedy <tjreedy@udel.edu>
bpo-39040: Fix parsing of email mime headers with whitespace between encoded-words. (gh-17620)
* bpo-39040: Fix parsing of email headers with encoded-words inside a quoted string.
It is fairly common to find malformed mime headers (especially content-disposition
headers) where the parameter values, instead of being encoded to RFC
standards, are "encoded" by doing RFC 2047 "encoded word" encoding, and
then enclosing the whole thing in quotes. The processing of these malformed
headers was incorrectly leaving the spaces between encoded words in the decoded
text (whitespace between adjacent encoded words is supposed to be stripped on
decoding). This changeset fixes the encoded word processing inside quoted strings
(bare-quoted-string) to do correct RFC 2047 decoding by stripping that
whitespace.
(cherry picked from commit 21017ed904f734be9f195ae1274eb81426a9e776)
[3.9] bpo-40777: Initialize PyDateTime_IsoCalendarDateType.tp_base at run-time (GH-20493) (GH-20495)
Recent changes to _datetimemodule broke compilation on mingw; see the comments in this change for details.
FWIW, @corona10: this issue is why `PyType_FromModuleAndSpec` & friends take the `bases` argument at run time.
(cherry picked from commit 459acc551656785bc4a3363d65c7a60f822da8e3)
bpo-1294959: Try to clarify the meaning of platlibdir (GH-20332)
Try to make the meaning of platlibdir clear. The previous wording could
be misinterpreted to suggest that it will be used to find all shared
libraries on the system, and not just Python extensions. Furthermore,
it was unclear whether it affects third-party (site-packages) extensions
or not. The new wording tries to make its dual purpose clear,
and provide the additional example of extensions in site-packages.
(cherry picked from commit 242d95659b6b4ff4fb54b58a30454dafa311d4e9)
Skip new "racing" socket tests which fail randomly until someone fix
them, to ease analysis of buildbot failures (skip tests which are
known to be broken/unstable).
(cherry picked from commit 84ee7e1573d166fe7a9be676813e12523b62ab24)
Co-authored-by: Victor Stinner <vstinner@python.org>
Christian Heimes [Thu, 28 May 2020 12:09:38 +0000 (14:09 +0200)]
[3.9] bpo-40791: Use CRYPTO_memcmp() for compare_digest (GH-20456) (GH-20461)
hashlib.compare_digest uses OpenSSL's CRYPTO_memcmp() function
when OpenSSL is available.
Note: The _operator module is a builtin module. I don't want to add
libcrypto dependency to libpython. Therefore I duplicated the wrapper
function and added a copy to _hashopenssl.c..
(cherry picked from commit db5aed931f8a617f7b63e773f62db468fe9c5ca1)
Co-authored-by: Christian Heimes <christian@python.org>
* Use a more universal explanation of string interpolation rather than specifically referencing sprintf(), which depends on the reader having a C background.
bpo-40795: ctypes calls unraisablehook with an exception (GH-20452)
If ctypes fails to convert the result of a callback or if a ctypes
callback function raises an exception, sys.unraisablehook is now
called with an exception set. Previously, the error was logged into
stderr by PyErr_Print().
(cherry picked from commit 10228bad0452d94e66c964b625a0b61befa08e59)
Co-authored-by: Victor Stinner <vstinner@python.org>
```
D:\a\cpython\cpython\Modules\_zoneinfo.c(903,52): warning C4267: '=': conversion from 'size_t' to 'unsigned int', possible loss of data [D:\a\cpython\cpython\PCbuild\_zoneinfo.vcxproj]
D:\a\cpython\cpython\Modules\_zoneinfo.c(904,44): warning C4267: '=': conversion from 'size_t' to 'unsigned int', possible loss of data [D:\a\cpython\cpython\PCbuild\_zoneinfo.vcxproj]
D:\a\cpython\cpython\Modules\_zoneinfo.c(1772,31): warning C4244: '=': conversion from 'ssize_t' to 'uint8_t', possible loss of data [D:\a\cpython\cpython\PCbuild\_zoneinfo.vcxproj]
```
(cherry picked from commit e4799b95945b44eb0e2eea26473db8e0a49ed0ee)
Co-authored-by: Pablo Galindo <Pablogsal@gmail.com>
bpo-13097: ctypes: limit callback to 1024 arguments (GH-19914)
ctypes now raises an ArgumentError when a callback
is invoked with more than 1024 arguments.
The ctypes module allocates arguments on the stack in
ctypes_callproc() using alloca(), which is problematic
when large numbers of arguments are passed. Instead
of a stack overflow, this commit raises an ArgumentError
if more than 1024 parameters are passed.
(cherry picked from commit 29a1384c040d39659e7d01f1fd7b6eb71ef2634e)
Co-authored-by: Sean Gillespie <sean@swgillespie.me>
projects / thirdparty / Python / cpython.git / log
