The default behaviour of idlefork idle is to open an editor window instead
of a shell. Complex expressions may be run in a fresh environment by
selecting "run".  There are times, however, when a shell is desired.
Though one can be started by "idle -t 'foo'", this script is more
convenient.  In addition, a shell and an editor window can be started
in parallel by "idles -e foo.py".

pythonrun.c, run_pyc_file():  repair semantic error wrt CO_GENERATOR vs
CO_GENERATOR_ALLOWED.

that info to code dynamically compiled *by* code compiled with generators
enabled.  Doesn't yet work because there's still no way to tell the parser
that "yield" is OK (unlike nested_scopes, the parser has its fingers in
this too).
Replaced PyEval_GetNestedScopes by a more-general
PyEval_MergeCompilerFlags.  Perhaps I should not have?  I doubted it was
*intended* to be part of the public API, so just did.

* iterators
* generators
* copied the nested scopes section from the 2.1 article
* standard library changes

the yield statement.  I figure we have to have this in before I can
release 2.2a1 on Wednesday.

Note: test_generators is currently broken, I'm counting on Tim to fix
this.

Others:  Remove redundant includes of assert.h.

   Include sys/poll.h if it was found by the configure script.  The OpenGroup
   spec says poll.h is the correct header file to use, so that file is
   preferred.

   newwin() wrapper to always return a window, and never a pad.  This makes
   the code match the documentation.

   by adding the contents of CCSHARED to the compiler specified by CC

code only compiled in debug mode, and I dutifully comply.

It's sort-of starting to work, but there's still problems with redraws and
with resizing the window.

long_format can reuse a scratch area for its repeated divisions (instead
of malloc/free for every digit produced); speeds str(long)/repr(long).

this module slightly more resiliant in the face of XHTML input, or just
colons in attribute names.

copy the Idle help.txt file.

Ref SF Python Patch 422471

"Make copy, cut and paste events case insensitive.  Reported by Patrick
K. O'Brien on idle-dev. (Should other bindings follow suit?)" --GvR

"Move the action of loading the configuration to the IdleConf module
rather than the idle.py script.  This has advantages and
disadvantages; the biggest advantage being that we can more easily
have an alternative main program."  --GvR

"Quick update to the extension mechanism (extend.py is gone, long live
config.txt)" --GvR

"Refactored, with some future plans in mind. This now uses the new
gotofileline() method defined in FileList.py"  --GvR

"Amazing.  A very subtle change in policy in descr-branch actually
found a bug here.  Here's the deal: Class PyShell derives from class
OutputWindow.  Method PyShell.close()
wants to invoke its parent method, but because PyShell long ago was
inherited from class PyShellEditorWindow, it invokes
PyShelEditorWindow.close(self).  Now, class PyShellEditorWindow itself
derives from class OutputWindow, and inherits the close() method from
there without overriding it.  Under the old rules,
PyShellEditorWindow.close would return an unbound method restricted to
the class that defined the implementation of close(), which was
OutputWindow.close.  Under the new rules, the unbound method is
restricted to the class whose method was requested, that is
PyShellEditorWindow, and this was correctly trapped as an error." --GvR

"Taught IDLE's autoident parser that "yield" is a keyword that begins a
stmt.  Along w/ the preceding change to keyword.py, making all this
work w/ a future-stmt just looks harder and harder." --tim_one

(From Rel 1.8: "Hack to make this still work with Python 1.5.2.  ;-( "
--fdrake)

"Move the action of loading the configuration to the IdleConf module
rather than the idle.py script.  This has advantages and
disadvantages; the biggest advantage being that we can more easily
have an alternative main program." --GvR

"Delete goodname() method, which is unused. Add gotofileline(), a
convenience method which I intend to use in a
variant. Rename test() to _test()."  --GvR

This was an interesting merge. The join completely missed removing
goodname(), which was adjacent, but outside of, a small conflict.
I only caught it by comparing the 1.1.3.2/1.1.3.3 diff.  CVS ain't
infallible.

"Remove legacy support for the BrowserControl module; the webbrowser
module has been included since Python 2.0, and that is the preferred
interface." --fdrake

special character bite us again.  ;-(

This fixes SF bug #440911.