Doc/build/html/library/idle.html, which is built from Doc/library/idle.rst.
Help.html is an rstrip()ed copy of idle.html that can be pushed.

The displayed content is unchanged because a) the version number is not
displayed, being inherent in the version displaying the file and b) the tt
tag was ignored by the custom parser in idlelib/help.py.  The wrapped
<span class="pre"> tags remain as they were.

This will be merged in 3.6 (without conflict) because the 3.6 help.html is
a copy of the 3.5 file (which was a copy of the 3.4 file).  The two files will
remain the same until either a) there is a content change that only applies
to one of them or b) 3.5 maintenance ends and the 3.6 file is itself rebased.

Closes #25910: fix dead and permanently redirected links in the docs. Thanks to SilentGhost for the patch.

Patch from Chi Hsuan Yen.

Based on patch by Daan Bakker.

This is a regression introduced in 3.5 by revision eba80326ba53. Fix by Jelte
Fennema, test case by me.

Patch by SilentGhost.

and link with a private copy of OpenSSL, like installers targeted
for 10.5 already do, since Apple has deprecated use of the system
OpenSSL and removed its header files from the Xcode 7 SDK.  Note
that this configuration is not currently used to build any
python.org-supplied installers and that the private copy of
OpenSSL requires its own root certificates.

As of Xcode 7, SDKs for Apple platforms now include textual-format stub
libraries whose file names have a .tbd extension rather than the
standard OS X .dylib extension.  The Apple compiler tool chain handles
these stub libraries transparently and the installed system shared libraries
are still .dylibs.  However, the new stub libraries cause problems for
third-party programs that support building with Apple SDKs and make
build-time decisions based on the presence or paths of system-supplied
shared libraries in the SDK.  In particular, building Python itself with
an SDK fails to find system-supplied libraries during setup.py's build of
standard library extension modules.  The solution is to have
find_library_file() in Distutils search for .tbd files, along with
the existing types (.a, .so, and .dylib).  Patch by Tim Smith.

Issue #26302: Correctly identify comma as an invalid character for a cookie (correcting regression in Python 3.5).

for issues #25995 and #25911.

Patch by Swati Jaiswal.

* There are only two base-64 alphabets defined by the RFCs, not three
* Due to the internal translation, plus (+) and slash (/) are never discarded
* standard_ and urlsafe_b64decode() discard characters as well

Also update the doc strings to clarify data types, based on revision
92760d2edc9e, correct the exception raised by b16decode(), and correct the
parameter name for the base-85 functions.

current versions of OpenBSD and NetBSD.  Patch by A. Jesse Jiryu Davis.

It can now handle OpenSSL versions 1.0.2e and greater, which don't
include include files in include/.

Note that sources prepared by this script no longer support the old
project files for 2.7; you now have to have Perl available to use
the old build_ssl.py script with sources from svn.python.org.

* http.client.HTTP (does not exist in Python 3) → HTTPConnection
* Server (deprecated) → ServerProxy
* Transport.send_request() grew a new “debug” parameter in Python 3 (revision
  a528f7f8f97a)

After 25032ec29315, compile() will raise a ValueError
if source contains null bytes.

Patch by SilentGhost.

importlib.util.LazyLoader.

The class was checking its argument as to whether its implementation
of create_module() came directly from importlib.abc.Loader. The
problem is that the classes coming from imoprtlib.machinery do not
directly inherit from the ABC as they come from _frozen_importlib.
Because the documentation has always said that create_module() was
ignored, the check has simply been removed.

'level' is specified but no __package__.

This brings the function inline with builtins.__import__(). Thanks to
Manuel Jacob for the patch.

* The Windows-specific binary notice was probably a Python 2 thing
* Make it more obvious gettarinfo() is based on stat(), and that non-ordinary
  files may need special care
* The file name must be text; suggest dummy arcname as a workaround
* Indicate TarInfo may be used directly, not just via gettarinfo()