trying to load an extension module.

While at it, also add a proper unittest.skipIf() guard to another test
involving imp.load_dynamic().

descriptor supported.

than the one merged from 3.2 to match the actual changes here.

libffi to 3.0.13.  (i'm not sure if it is needed anymore but see
issue 10309 for details which makes no mention of upstream; this
change is already in 3.3 and 3.4 but may need reapplying to 2.7
as done here)

  were missing from a messed up merge during the libffi 3.0.13
  import.  the diffs from upstream libffi 3.0.13 are now small.

  that was undone in the merge fun from upstream which already
  had it in 3.0.13.
* Add the missing update to libffi.info.

  in the merge fun from upstream which already had it in 3.0.13.
* Add the missing update to libffi.info.

raise an ImportError (rather than fatal) when __import__ is not found in __builtins__ (closes #17867)

modules which will typically work when reloaded.

A hack is used to support these functions as their API allowed them to
pass in a file object but then operate as if import had loaded them.
Unfortunately the hack kept a reference around for the file object
passed in which would be unusable on reload since it had been closed.
The solution is to simply use the hack for the initial load but then a
proper loader on the module so that imp.reload() at least has a chance
to work.

I know one of them looks worse, but now it is <80 chars and
matches the 2.7 text.

Behavior confirmation and initial patch by Jessica McKellar.

Original patch by Kushal Das.

such as was shipped with Centos 5 and Mac OS X 10.4.

importlib.invalidate_caches().

Without this changeset plistlib would write empty tags for plistlib.Data
objects in deeply nested datastructures.

Fixes #17353

condition variable.  Original patch by Thomas Rachel.

Patch by Jonas Wagner