(Contributed by Walter Dörwald).

* Convert three test modules to unittest format.
* Expanded coverage in test_structseq.py.
* Raymond added a new test in test_sets.py

When the indents were set to longer than the width and long word breaking
was enabled, an infinite loop would result because the inner loop did not
assure that at least one character was stripped off on every pass.

Avoid Armin Rigo's dastardly exercise in re-entrancy.

Rework previous fix slightly; the &0x20 test seems useless, and the isprint() check mustn't prevent the meta-bit check at the end

[Bug #776542] open_https() generates a bad Authorization header because it calls .putheader() wrongly.  Reported by Steffen Ries.

   actually occurred in 2.2.  Move the description to whatsnew22.tex.

[Bugfix candidate -- I'll backport to 2.3 (another 2.2 bugfix seems unlikely)

Closes bug #796149 .  Will be backported.

database environments to use shared memory on systems supporting it.

Make sure the inner function is not compiled when there is a syntax
error in the default arguments.

Bug fix candidate.

Will backport to 2.3 myself.

Adding missing support for '%F'.

Will backport to 2.3.1.

platforms (e.g., Cygwin) that are "particular" about open files, this will
cause other regression tests that use the same temp file to fail:

    $ ./python.exe -E -tt Lib/test/regrtest.py -l
    test_largefile test_mmap test_mutants
    test_largefile
    test test_largefile failed -- got -1794967295L, but expected 2500000001L
    test_mmap
    test test_mmap crashed -- exceptions.IOError: [Errno 13] Permission denied: '@test'
    test_mutants
    test test_mutants crashed -- exceptions.IOError: [Errno 13] Permission denied: '@test'

This patch solves the problem by adding missing "try/finally" blocks. Note
that the "large" size of this patch is due to many white space changes --
otherwise, the patch is small.

I tested this patch under Red Hat Linux 8.0 too.

The -- is special to TeX and was printing as just -.

Thou shalt not count to two unless proceeding to the number three.

Python 1.5.2 wrong in the section on config files.

I have a different WISE version, and it's installed in a different directory.
My sytem directory is somewhere else.

test_rfc2231_no_language_or_charset_in_boundary(),
test_rfc2231_no_language_or_charset_in_charset(): New tests for proper
decoding of some RFC 2231 headers.

Backport candidate (as was the Utils.py 1.25 change) to both Python
2.3.1 and 2.2.4 -- will do momentarily.

can be None, and what to do in that situation.

get_filename(), get_boundary(), get_content_charset(): Make sure these
handle RFC 2231 headers without a CHARSET field.

Backport candidate (as was the Utils.py 1.25 change) to both Python
2.3.1 and 2.2.4 -- will do momentarily.

charset and language fields are not present, e.g. as in:

    title*0="This%20is%20encoded"

which is a disaster on Windows.  Restored the binary default of all
previous releases.  Also minor code cleanups.

Bugfix candidate!

NULL pointer. (Detected by Michael Hudson, patch provided by Neal Norwitz).

Fix refcounting leak in filtertuple().

Will backport to release23-maint.