Fix my previous commit (r85569).

filenames encoded to the filesystem encoding with surrogateescape error handler
(to support undecodable bytes), instead of UTF-8 in strict mode.

It occurs if the path contains surrogates.

* Expose the build flags to Python as sys.abiflags
* Shared library libpythonX.Y<abiflags>.so
* python-config --abiflags
* Make two distutils tests that failed with --enable-shared (even before this
  patch) succeed.
* Fix a few small style issues.

Patch by Sébastien Sablé.

filesystem encoding instead of utf-8.

imp_cache_from_source() encodes the input path to filesystem encoding and this
path is passed to make_compiled_pathname().

path

module path

importing encoded modules that do not depend on specific stdlib
modules being encoded in a certain way.

#9683: remove broken dead code dealing with nested arguments removed from Py3k, and update the docs and docstrings accordingly.

#2830: add html.escape() helper and move cgi.escape() uses in the standard library to it.  It defaults to quote=True and also escapes single quotes, which makes casual use safer.  The cgi.escape() interface is not touched, but emits a (silent) PendingDeprecationWarning.

#5355 followup: add unit test for new dictionaries, and provide submodules from xml.parsers.expat as advertised.

#5355: Provide mappings from Expat error numbers to string descriptions and backwards, in order to actually make it possible to analyze error codes provided by ExpatError.

Amaury noticed that we're not waiting for the subprocess to be ready --
it should be checking for 1, not 0.

Sorry for the ephemeral inconvenience.