The contents of this attribute are an implementation detail, as
documented for #9442, so we should not parse it, to support non-CPython
VMs in the future.

Unfortunately, one use comes directly from PEP 345, so an edit will have
to be agreed before fixing the code (see comment in d2.markers).

Other remaining uses are found in d2.compiler and could be replaced by
the platform module (which also parses sys.version, but then it wouldn?t
be my fault :)

I broke test_mixin2to3 somehow; distutils2-default is okay and packaging too,
so I don?t see an obvious reason right now, I?ll investigate later.

All code (util.byte_compile, build_py, install_lib) can now create .pyc
and/or.pyo files according to options given by users, without
interference from the calling Python?s own optimize mode or from the
sys.dont_write_bytecode switch.

The rationale is that packaging gives control over the creation of
.pyc/.pyo files to the user with its own explicit option, and the
behavior should not be changed if the calling Python happens to run with
-B or -O for whatever reason.

This is actually a bug fix, not an improvement: Digging into the early
history of distutils shows that the original author wanted this behavior
(see for example comments in build_py in r12940).

- Use os.makedirs (I had forgotten about it!)
- Let TempdirManager.write_file call os.path.join for us
- Let TempdirManager.tearDown go back to the previous working directory
  for us
- Use a skip instead of hiding a method with an underscore
- Reset the finalized attribute of command objects before calling
  ensure_finalized a second time, so that it?s not a no-op
- Address pyflakes warnings

This codebase is compatible with 3.1, 3.2 and 3.3.  It was converted with 2to3
and a semi-automated diff/merge with packaging in 3.3 to fix some idioms.
We?ve now come full circle from 2.x to 3.x to 2.x to 3.x again :)

Starting from now, contributors can make patches for packaging (preferred, as
the stdlib?s regrtest is very useful), distutils2 or distutils-python3, and
we?ll make patches flow between versions.

The existing test_record is not easily extendable to add script files or
extension modules: it collects all files from fake_dists and generates a
RECORD file at runtime.  I felt more comfortable adding a new test
written from scratch, more self-contained (just one project with
well-defined files) and more stupid (the checksums and sizes are
computed once and hard-coded).

(i.e. to check the files found are what we expect)