I didn't realize LC_ALL was an override, and I should have.  I tried to
make a test, but it is not clear that the LC variables actually affect
the strings that uuid is using to parse the command output.

locale-depending str.lower() in locale.normalize().

requires them.  Disable executable bits and shebang lines in test and
benchmark files in order to prevent using a random system python, and in
source files of modules which don't provide command line interface.

Patch from Andrew Barnert.

wsgiref.application_uri() and wsgiref.request_uri() functions.

Fixed regression inroduced in issue #6157.

Filename such as r'c:\users\db3l\appdata\local\temp\tmph3vkvf' contains '\t'
which is interpreted by ast.literal_eval() as a tabulation.

ifconfig executable is not available.

Incorrect requirements can cause memory swapping.

a help on environment with too small COLUMNS.  Based on patch by
Elazar Gershuni.

zip_searchorder scanning loop in get_module_code().
[already done in 3.3 and 3.4]

* Misc.image_names(), Misc.image_types(), Wm.wm_colormapwindows(), and
  LabelFrame.panes() now always return a tuple.
* Fixed _stringify() for non-ASCII strings.
* Fixed error of comparing str and int in tt.LabeledScale._adjust().
* ttk.Notebook.index() now always returns int.
* ttk.Notebook.tabs() now always returns a tuple.
* ttk.Entry.bbox() now always returns a tuple of ints.
* ttk.Entry.validate() now always correctly works.
* ttk.Combobox.current() now always returns int.
* ttk.Panedwindow.sashpos() now always returns int.
* ttk.Treeview.bbox() now always returns a tuple of ints.
* ttk.Treeview.get_children() now always returns a tuple.
* ttk.Treeview.exists() now always correctly works.
* ttk.Treeview.index() now always returns int.
* ttk.Treeview.tag_has() now always returns 0 or 1.
* And numerous other errors in methods which returns a tuple, list or dict.
* Fixed ttk tests for wantobjects is False.

posix module ImportError cause the nt module import to fail.

from is modified during the lifetime of the Python process after
zipimport has already opened and cached the zip's table of contents
it now fstat's the file after opening it upon every attempt to access
anything within and will re-read the table of contents if the .zip file
inode, size or mtime have changed.

It would've been nicer to hold any .zip file used by zipimport open for the
duration of the process but that would be more invasive and add an additional
open file descriptor to all zipimport using processes.  It also would likely
not fix the problem on Windows due to different filesystem semantics.