and "" in grep.getgrgid(). Adjust the test to work
around this problem. This should fix SF bug #724771.

Revised netrc.py to include the additional ascii punctuation
characters.  Omitted the other logic changes.  See
Lib/netrc.py 1.17.

Since this is more of a feature request than a bug,
including in Py2.3 but not recommending for backporting.

Expanded the range of allowable characters to include ascii punctuation.
Allows resource files to have a larger character set for passwords.
(Idea contributed by Bram Moolenaar.)

(contributed by logistix; substantially reworked by rhettinger).

To create a representation of non-string arrays, array_repr() was
starting with a base Python string object and repeatedly using +=
to concatenate the representation of individual objects.

Logistix had the idea to convert to an intermediate tuple form and
then join it all at once.  I took advantage of existing tools and
formed a list with array_tolist() and got its representation through
PyObject_Repr(v) which already has a fast implementation for lists.

This is used on www.python.org.

 - upgrade to version 0.4.8

docs here are best-guess:  the MS docs I could find weren't clear, and
some even claimed _commit() has no effect on Win32 systems (which is
easily shown to be false just by trying it).

Adding code to handle it properly.

Fix missing right parenthesis.
Add three examples.

management.  Old code still #ifdef'd out - I may remove this in a sec,
but for now, let's get it in and things passing the tests again!

field contains garbage after the first NUL character.

using a build directory just inside the source directory and saving
just one copy of the test data in the source tree, rather than having
a copy in each build directory.

accordingly to further discussions with Greg Chapman in patch #712900.

directory.

- various adjustments to sum() markup and explanation

Fixes 719303.

were indeed files that weren't unpacked.

and group names. This should fix SF bug #724771.

* Can now test for basic blocks.
* Optimize inverted comparisions.
* Optimize unary_not followed by a conditional jump.
* Added a new opcode, NOP, to keep code size constant.
* Applied NOP to previous transformations where appropriate.

Note, the NOP would not be necessary if other functions were
added to re-target jump addresses and update the co_lnotab mapping.
That would yield slightly faster and cleaner bytecode at the
expense of optimizer simplicity and of keeping it decoupled
from the line-numbering structure.

  * UINT_MAX -> ULONG_MAX since we are dealing with longs
  * ParseTuple needs &int for 'i' and &long for 'l'
    There may be a better way to do this, but this works.

Remove "." in coverage output for lines with comments.
Silence complaints in coverage output about unexecuted docstrings.

Eliminate use of inspect module in favor of direct access to frame and
code objects.  We're in a trace function here:  Efficiency counts!

Remove unused code.  Reflow long lines.

Remove backwards compatibility for stored trace output from Zooko's
experiment to add calledfuncs to the pickled dict.

Move code to generate per-file coverage stats to a separate routine.

Replace use of parser module with call to compile.

- restructure build for modules now in Python DLL

README.os2emx
- clean out old cruft no longer appropriate now that EMX port builds
  from CVS
- reflect move of modules into core DLL
- add section on building from source

- add _csv module to the build list
- various cleanups

config.c:
- various cleanups

pyconfig.h:
- various cleanups