This fixes some of the failures, but there are still more.

This test now passes after removing refs to __coerce__.

Highlighting tab indents is off at the moment since that would be rather noisy,
but should be turned on in the future when conversion to the new indentation
style has been done.

Change file.readinto() to require binary mode.

Get rid of silly lambdas in the unit test suite.
Add a TODO list to the unit test suite (TDD style).

(Plus, it fails. :-)

This resolves a problem found by Thomas Wouters:
http://mail.python.org/pipermail/python-dev/2006-April/064051.html

I really don't know why I bother; these are all generated files.
But I don't recall how to regenerate them nor how to fix the generator.

The hardest part was fixing two mutual recursive imports;
somehow changing "import foo" into "from . import foo" where
foo and bar import each other AND both are imported from __init__.py
caused things to break.  Bah.

I'm not convinced this is the right thing to do...  InstanceType makes no sense any longer.

number of tests, all because of the codecs/_multibytecodecs issue described
here (it's not a Py3K issue, just something Py3K discovers):
http://mail.python.org/pipermail/python-dev/2006-April/064051.html

Hye-Shik Chang promised to look for a fix, so no need to fix it here. The
tests that are expected to break are:

test_codecencodings_cn
test_codecencodings_hk
test_codecencodings_jp
test_codecencodings_kr
test_codecencodings_tw
test_codecs
test_multibytecodec

This merge fixes an actual test failure (test_weakref) in this branch,
though, so I believe merging is the right thing to do anyway.

usable again.

state (unless you happen to have a re.pyc lying around), but it'll be fixed
in the next merge.

(branch-creation time) up to 43067. 43068 and 43069 contain a little
swapping action between re.py and sre.py, and this mightily confuses svn
merge, so later changes are going in separately.

This merge should break no additional tests.

The last-merged revision is going in a 'last_merge' property on '.' (the
branch directory.) Arbitrarily chosen, really; if there's a BCP for this, I
couldn't find it, but we can easily change it afterwards ;)

it's fundamentally unmaintainable.  Should we throw away pyclbr, or rewrite
it using the AST?

API.

But shouldn't we kill this module?  How many pprint clones do we need?

(There are some other problems with test_class.py that aren't as
easily fixed. :-( )

- a line indented with tabs;
- a function named 'as'.

silently converts floats to integers when expecting integer values.

'dictproxy' (which is a read-only non-dict mapping type that can't be passed
to exec.)

The failures the test finds are behavioural differences between old- and
new-style classes that may or may not be intended.

expected, when dealing with new-style broken-iterators.

convert float arguments to integer-taking format characters, so fix the test
to expect the failure.

 - Expect a new-style class tree in the getclasstree test.

sooner for new-style broken-iterators, expect it to.

Add (equally superficial) >>=/<<= test in the process. Relies on floats that
should be extremely close to the int '6' printing as '6.0', but I believe
that's a valid assumption ;P

Since the broken iterators are now new-style classes, iter() was able to do
the valid-iterator check sooner (on instantiation instead of on first call),
making the tests blow up sooner than expected.

(New-style class behaviour was already thoroughly tested)

   printing differently.
 - Fix doctest for classic-class behaviour, make it test new-style behaviour
   on an implicitly-new-style class instead.