test, make sure to check if __builtins__ is a dict or not.

Discovered when running importlib.test.regrtest.

filenames and enable os.fsencode().

Note: file system encoding cannot be None anymore (since r81190, issue #8610).

 * replace "the default encoding" by "'utf-8' encoding"
 * fix "w" / "w*" / "w#" doc: similar to "y" / "y*" / "y#"
   and not "s" / "s*" / "s#"
 * "u#": remove "Non-Unicode objects are handled by interpreting their
   read-buffer pointer ...", it's no more true
 * "es", "es#": remove "... and objects convertible to Unicode into a character
   buffer", it's no more true
 * Py_BuildValue(), "K" and "L" formats: specify the name of the C type on
   Windows (_int64 / unsigned _int64) as done for PyArg_Parse*() long long
   types
--CETTE ligne, et les suivantes ci-dessous, seront ignorées--

M    Doc/c-api/arg.rst

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81906 | benjamin.peterson | 2010-06-11 16:40:37 -0500 (Fri, 11 Jun 2010) | 1 line

  different spellings are just unacceptable
........

value to bytes instead of str (as done for c_char and c_char_p).

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81907 | antoine.pitrou | 2010-06-11 23:42:26 +0200 (ven., 11 juin 2010) | 5 lines

  Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
  the interpreter with characters outside the Basic Multilingual Plane
  (higher than 0x10000).
........

........
  r81904 | mark.dickinson | 2010-06-11 21:27:05 +0100 (Fri, 11 Jun 2010) | 4 lines

  Fix possible undefined behaviour from signed overflow in struct module.

  Backport of revisions 81897, 81898 and 81902 from py3k.
........

........
  r81899 | victor.stinner | 2010-06-11 21:22:28 +0200 (ven., 11 juin 2010) | 2 lines

  Issue #8362: Add Misc/maintainers.rst: list of module maintainers
........

objects (Fraction, float, complex, int) now all function as expected.

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81889 | ezio.melotti | 2010-06-11 05:21:25 +0300 (Fri, 11 Jun 2010) | 1 line

  Remove extra ] from itertools.count docstring.
........

File system encoding have to be hardcoded to "utf-8" on Mac OS X. r81190
introduced a regression: the encoding was changed depending on the locale.

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81878 | michael.foord | 2010-06-10 21:40:21 +0100 (Thu, 10 Jun 2010) | 2 lines

  Fix issue with nested test suites debug method and module setups. (unittest)
........

........
  r81876 | michael.foord | 2010-06-10 17:32:00 +0100 (Thu, 10 Jun 2010) | 1 line

  NEWS update for issue 8948.
........

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81859 | michael.foord | 2010-06-09 13:29:56 +0100 (Wed, 09 Jun 2010) | 1 line

  Typo correction.
........

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81853 | michael.foord | 2010-06-08 23:44:52 +0100 (Tue, 08 Jun 2010) | 1 line

  Issue 8948. cleanup functions are not run by unittest.TestCase.debug(), plus class and module teardowns are not run by unittest.TestSuite.debug().
........

instead of warning.  This makes it consistent with the other integer
codes.

Don't use normalize_encoding() result if it is truncated.

enable shortcuts for upper case encoding name. Add also a shortcut for
"iso-8859-1" in PyUnicode_AsEncodedString() and PyUnicode_Decode().

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81864 | alexander.belopolsky | 2010-06-09 13:08:11 -0400 (Wed, 09 Jun 2010) | 1 line

  Fixed markup of tm_isdst attribute.
........

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81860 | antoine.pitrou | 2010-06-09 18:24:00 +0200 (mer., 09 juin 2010) | 3 lines

  Issue #8930: fix some C code indentation
........

Issue minidom.unlink with a context manager

Remove last references to the "char buffer" of the buffer protocol from
Python3.

instead of ignoring the error and leave the pointer to the string and the size
unchanged (not initialized).

Fix also the type in the error message of "Z", "Z#" and "Y" formats.

Call PyErr_PrintEx(0) instead of PyErr_Print() to avoid a crash if
Py_FatalError() is called in an early stage of Python initialization (if PySys
is not yet initialized).

variable declaration outside of a loop. #2810 was when this first went in.

mywrite() falls back to the C file object if sys_pyfile_write() returns an
error. This patch fixes a segfault is Py_FatalError() is called in an early
stage of Python initialization.