The Windows build now depends on Python 3.6 to fetch externals, but it will be downloaded via NuGet (which is downloaded via PowerShell) if it is not available via `py -3.6`.  This means the only thing that must be installed on a modern Windows box to do a full build of CPython with all extensions is Visual Studio.

Also fixes an outdated note about _lzma in PCbuild/readme.txt

* test dedent with declining indent level
* add textwrap.dedent test cases

In rare circumstances PyImport_Import() could return NULL without raising
an error.

When os.spawnv() fails while handling arguments, free correctly
argvlist: pass lastarg+1 rather than lastarg to free_string_array()
to also free the first item.

In the C locale on Mac OS X, the default filesystem encoding
used for operating system interfaces is UTF-8, but the
default encoding used on the standard streams is still ASCII.

Setting the POSIX locale also behaves differently from setting
other locales on Mac OS X, so skip that in the test suite for now.

* bpo-29591: Upgrade Modules/expat to libexpat 2.2

* bpo-29591: Restore Python changes on expat

* bpo-29591: Remove expat config of unsupported platforms

Remove the configuration (Modules/expat/*config.h) of unsupported
platforms:

* Amiga
* MacOS Classic on PPC32
* Open Watcom

* bpo-29591: Remove useless XML_HAS_SET_HASH_SALT

The XML_HAS_SET_HASH_SALT define of Modules/expat/expat.h became
useless since our local expat copy was upgrade to expat 2.1 (it's now
expat 2.2.0).

Use a build/ directory in the build directory, not in the source
directory, since the source directory may be read-only and must not
be modified.

Fallback on the source directory if the build directory is not
available (missing "abs_builddir" sysconfig variable).

bpo-30436: Raise ModuleNotFoundError for importlib.util.find_spec() when parent isn't a package (GH-1899)

Previously AttributeError was raised, but that's not very reflective of the fact that the requested module can't be found since the specified parent isn't actually a package.

Before, <return> would not, for instance, complete 're.c' to 're.compile' even with 'compile' highlighted.  Now it does.  Before, <return> was inserted into text, which in Shell meant compile() and possibly execute.  Now cursor is left after completion.

The public cyrus.andrew.cmu.edu IMAP server (port 993) doesn't accept
TLS connection using our self-signed x509 certificate. Remove the two
tests which are already skipped.

Write a new test_certfile_arg_warn() unit test for the certfile
deprecation warning.

The root problem was non-check for hide_event. When user clicks
on autocomplete window (acw), root widget gets focusOut event, then triggers hide_window to
close the acw. It should only be hide when acw is active, and acw didn't
get focus at FocusOut event (this event bind on acw and widget), or when
widget get a ButtonPress event (this event only bind on widget).

MacOS  froze after double click on acw because when
doubleclick_event try to hide window at the end, hide_window function
destory whole acw, but tkinter didn't get focus back to widget. So set focus
on widget first, then destory acw.

Windows could not respond on double click event, because of the misbehavior
of Configure event.  When acw was shown, tkinter called winconfig
event multiple times.  That caused tkinter to not response to double
click event.  When on Windows, unbind Configure event first
time get into winconfig_event to prevent multiple call of this event.

Replace platform with platforms.

When os.spawnve() fails while handling arguments, free correctly
argvlist: pass lastarg+1 rather than lastarg to free_string_array()
to also free the first item.

PPC64 Fedora 3.x buildbot requires at least a delta of 14 ms: revert
the utime delta to 20 ms.

I tried 10 ms, but test_os failed on the PPC64 Fedora 3.x buildbot.

When IDLE fail to start because the socket connection fails, direct people to a new subsection of the IDLE doc listing various causes and remedies.

_test_multiprocessing.test_queue_feeder_donot_stop_onexc() now uses a
timeout of 1 second on Queue.get(), instead of 0.1 second, for slow
buildbots.

On Windows, tolerate a delta of 50 ms instead of 20 ms in
test_utime_current() and test_utime_current_old() of test_os.

On other platforms, reduce the delta from 20 ms to 10 ms.

For unknown reasons, this does not work when running leak tests.

`PyModule_New()` now refers to `PyModule_NewObject()`

bpo-24744: Raise error in pkgutil.walk_packages if path is str

Previously an empty result list was accidentallly returned, since the
code iterated over the string as if it were the expected list of paths,
and of course found nothing.

* bpo-24484: Avoid race condition in multiprocessing cleanup

The finalizer registry can be mutated while inspected by multiprocessing
at process exit.

* Use test.support.start_threads()

* Add Misc/NEWS

Patch by Louie Lu.

When checking for reference leaks, test_c_locale_coercion is run
multiple times and so _LocaleCoercionTargetsTestCase.setUpClass() is
called multiple times. setUpClass() appends new value at each call,
so it looks like a reference leak.

Moving the setup from setUpClass() to setUpModule() avoids
this, eliminating the false alarm.

Don't run Python in an empty environment, but copy the current
environment and set PYTHONIOENCODING. So the test works also on
Python compiled in shared mode (using libpython).

* Fix race condition in signal wakeup in forkserver (followup to PR #1989)

There's an admittedly well-known race condition where ECHILD can arrive
just before the C function epoll_wait() and the latter wouldn't therefore
return EINTR.  The solution is to use set_wakeup_fd(), which was designed
to avoid such race conditions.

* Reset wakeup fd in child

Move up the discussion about 'with' keyword, so it appears earlier in the document.

* Fix bpo-30589: improve Process.exitcode with forkserver

When the child is killed, Process.exitcode should return -signum, not 255.

* Add Misc/NEWS

bpo-30624 / selectors: use bare 'except' clause instead of 'except
Exception'