recognized by the code generator and code generation for the test and
the subsequent suite is suppressed.

One must write *exactly* ``if __debug__:'' or ``elif __debug__:'' --
no parentheses or operators must be present, or the optimization is
not carried through.  Whitespace doesn't matter.  Other uses of
__debug__ will find __debug__ defined as 0 or 1 in the __builtin__
module.

This is done for backward compatibility with Python 1.4.

this mode, all tests are run in verbose mode with their output to
stdout.  No comparing of output is done.

of day, day of week, and season.

Fix the weekday predictions -- these seemed to be all bogus.  The new
predictions seem to correspond with strftime() on Solaris and IRIX, so
I believe they are correct.

Get rid of the test for non-standard format %C returning "the same as
date(1)".  This is hard to do reliably without opening a pipe to date,
and moreover, on IRIX 6.2, %C yields the Century.  So we use that
instead.  (We don't complain about this in non-verbose mode anyway.)

integer overflow in << operator.

it anyway.

Fred: sorry, I hadn't checked these changes in.  This should fix your
tracebacks!

nonzero.

on-line list of same.

- Changed FragLoader.h to CodeFragments.h
- Removed Desk.h

- Changed FragLoader.h to CodeFragments.h
- Removed Desk.h
- Regenerated bgen modules from new universal headers
- Changed some of the s# in PyArg_ParseTuple to m# (unfortunately:
  this should have been a different commit)

and there is already a similar demo in Demo/classes/Complex.py.

Lee is wondering whether to withdraw his patchs.  Sigh.

extensively in test_strftime.py anyway).

and there, etc. These projects are still CW10.

  in importdl.c (I had just one crash too many with a static python
  importing a dynamic module)
- On powerpc, enable USE_CACHE_ALIGNED with a linesize of 32 bytes.