"from blah import (foo, bar
baz, bongo)"

ldexp.  Both methods are exact, and return the same results.  Turns out
multiplication is a few (but just a few) percent faster on my box.
They're both significantly faster than using struct with a Q format
to convert bytes to a 64-bit long (struct.unpack() appears to lose due
to the tuple creation/teardown overhead), and calling _hexlify is
significantly faster than doing bytes.encode('hex').  So we appear to
have hit a local minimum (wrt speed) here.

components without division and without roundoff error for properly
sized mantissas (i.e. on systems with 53 or more mantissa bits per
float).  Eliminates the previous implementation's rounding bias as
aptly demonstrated by Tim Peters.

system, exploit cmd.exe's setlocal function in this directory.

Replaced outcomes from native Tcl/Tk tests.  Maybe the diffs are legit,
maybe not.  I noticed that the Tcl results I'm replacing here claimed
both that there were no failures, and that one file had tests with
failures, so I wasn't inclined to trust them <wink>.

don't.

error msgs.

in first column, no parens around return value).

to unittest, so make it official:  new module constants COMPARISON_FLAGS
and REPORTING_FLAGS, which are bitmasks or'ing together the relevant
individual option flags.

set_unittest_reportflags():  Reworked to use REPORTING_FLAGS, and
simplified overly complicated flag logic.

class FakeModule:  Removed this; neither documented nor used.

heuristics for filtering out imported names.

1) When a breakpoint is set via a function name:
- the breakpoint gets the lineno of the def statement
- a new funcname attribute is attached to the breakpoint

2) bdb.effective() calls new function checkfuncname() to handle:
- def statement is executed: don't break.
- a first executable line of a function with a breakpoint on the lineno of the
def statement is reached: break.

This fixes bugs 976878, 926369 and 875404. Thanks Ilya Sandler.

* Use it for seeding when it is available.
* Provide an alternate generator based on it.

Bugfix candidate -- although long_pow() is so different now I doubt a
patch would apply to 2.3.

This checkin is adapted from part 2 (of 3) of Trevor Perrin's patch set.

BACKWARD INCOMPATIBILITY:  SHIFT must now be divisible by 5.  AFAIK,
nobody will care.  long_pow() could be complicated to worm around that,
if necessary.

long_pow():
  - BUGFIX:  This leaked the base and power when the power was negative
    (and so the computation delegated to float pow).
  - Instead of doing right-to-left exponentiation, do left-to-right.  This
    is more efficient for small bases, which is the common case.
  - In addition, if the exponent is large (more than FIVEARY_CUTOFF
    digits), precompute [a**i % c for i in range(32)], and go left to
    right 5 bits at a time.
l_divmod():
  - The signature changed so that callers who don't want the quotient,
    or don't want the remainder, can pass NULL in the slot they don't
    want.  This saves them from having to declare a vrbl for unwanted
    stuff, and remembering to decref it.
long_mod(), long_div(), long_classic_div():
  - Adjust to new l_divmod() signature, and simplified as a result.

This checkin is adapted from part 1 (of 3) of Trevor Perrin's patch set.

x_mul()
  - sped a little by optimizing the C
  - sped a lot (~2X) if it's doing a square; note that long_pow() squares
    often
k_mul()
  - more cache-friendly now if it's doing a square
KARATSUBA_CUTOFF
  - boosted; gradeschool mult is quicker now, and it may have been too low
    for many platforms anyway
KARATSUBA_SQUARE_CUTOFF
  - new
  - since x_mul is a lot faster at squaring now, the point at which
    Karatsuba pays for squaring is much higher than for general mult

test depends on invisible trailing whitespace in .py files.  The author will
have to repair that.