Merged revisions 57221-57391 via svnmerge from

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

........
  r57227 | facundo.batista | 2007-08-20 17:16:21 -0700 (Mon, 20 Aug 2007) | 5 lines


  Catch ProtocolError exceptions and include the header information in
  test output (to make it easier to debug test failures caused by
  problems in the server). [GSoC - Alan McIntyre]
........
  r57229 | mark.hammond | 2007-08-20 18:04:47 -0700 (Mon, 20 Aug 2007) | 5 lines

  [ 1761786 ] distutils.util.get_platform() return value on 64bit Windows
  As discussed on distutils-sig: Allows the generated installer name on
  64bit Windows platforms to be different than the name generated for
  32bit Windows platforms.
........
  r57230 | mark.hammond | 2007-08-20 18:05:16 -0700 (Mon, 20 Aug 2007) | 5 lines

  [ 1761786 ] distutils.util.get_platform() return value on 64bit Windows
  As discussed on distutils-sig: Allows the generated installer name on
  64bit Windows platforms to be different than the name generated for
  32bit Windows platforms.
........
  r57253 | georg.brandl | 2007-08-20 23:01:18 -0700 (Mon, 20 Aug 2007) | 2 lines

  Demand version 2.5.1 since 2.5 has a bug with codecs.open context managers.
........
  r57254 | georg.brandl | 2007-08-20 23:03:43 -0700 (Mon, 20 Aug 2007) | 2 lines

  Revert accidental checkins from last commit.
........
  r57255 | georg.brandl | 2007-08-20 23:07:08 -0700 (Mon, 20 Aug 2007) | 2 lines

  Bug #1777160: mention explicitly that e.g. -1**2 is -1.
........
  r57256 | georg.brandl | 2007-08-20 23:12:19 -0700 (Mon, 20 Aug 2007) | 3 lines

  Bug #1777168: replace operator names "opa"... with "op1"... and mark everything up as literal,
  to enhance readability.
........
  r57259 | facundo.batista | 2007-08-21 09:57:18 -0700 (Tue, 21 Aug 2007) | 8 lines


  Added test for behavior of operations on an unconnected SMTP object,
  and tests for NOOP, RSET, and VRFY. Corrected typo in a comment for
  testNonnumericPort. Added a check for constructing SMTP objects when
  non-numeric ports are included in the host name. Derived a server from
  SMTPServer to test various ESMTP/SMTP capabilities. Check that a
  second HELO to DebuggingServer returns an error. [GSoC - Alan McIntyre]
........
  r57279 | skip.montanaro | 2007-08-22 12:02:16 -0700 (Wed, 22 Aug 2007) | 2 lines

  Note that BeOS is unsupported as of Python 2.6.
........
  r57280 | skip.montanaro | 2007-08-22 12:05:21 -0700 (Wed, 22 Aug 2007) | 1 line

  whoops - need to check in configure as well
........
  r57284 | alex.martelli | 2007-08-22 14:14:17 -0700 (Wed, 22 Aug 2007) | 5 lines

  Fix compile.c so that it records 0.0 and -0.0 as separate constants in a code
  object's co_consts tuple; add a test to show that the previous behavior (where
  these two constants were "collapsed" into one) causes serious malfunctioning.
........
  r57286 | gregory.p.smith | 2007-08-22 14:32:34 -0700 (Wed, 22 Aug 2007) | 3 lines

  stop leaving log.0000001 __db.00* and xxx.db turds in developer
  sandboxes when bsddb3 tests are run.
........
  r57301 | jeffrey.yasskin | 2007-08-22 16:14:27 -0700 (Wed, 22 Aug 2007) | 3 lines

  When setup.py fails to find the necessary bits to build some modules, have it
  print a slightly more informative message.
........
  r57320 | brett.cannon | 2007-08-23 07:53:17 -0700 (Thu, 23 Aug 2007) | 2 lines

  Make test_runpy re-entrant.
........
  r57324 | georg.brandl | 2007-08-23 10:54:11 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1768121: fix wrong/missing opcode docs.
........
  r57326 | georg.brandl | 2007-08-23 10:57:05 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1766421: "return code" vs. "status code".
........
  r57328 | georg.brandl | 2007-08-23 11:08:06 -0700 (Thu, 23 Aug 2007) | 2 lines

  Second half of #1752175: #ifdef out references to PyImport_DynLoadFiletab if HAVE_DYNAMIC_LOADING is not defined.
........
  r57331 | georg.brandl | 2007-08-23 11:11:33 -0700 (Thu, 23 Aug 2007) | 2 lines

  Use try-except-finally in contextlib.
........
  r57343 | georg.brandl | 2007-08-23 13:35:00 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1697820: document that the old slice protocol is still used by builtin types.
........
  r57345 | georg.brandl | 2007-08-23 13:40:01 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1573854: fix docs for sqlite3 cursor rowcount attr.
........
  r57347 | georg.brandl | 2007-08-23 13:50:23 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1694833: fix imp.find_module() docs wrt. packages.
........
  r57348 | georg.brandl | 2007-08-23 13:53:28 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1594966: fix misleading usage example
........
  r57349 | georg.brandl | 2007-08-23 13:55:44 -0700 (Thu, 23 Aug 2007) | 2 lines

  Clarify wording a bit.
........
  r57351 | georg.brandl | 2007-08-23 14:18:44 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1752332: httplib no longer uses socket.getaddrinfo().
........
  r57352 | georg.brandl | 2007-08-23 14:21:36 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1734111: document struct.Struct.size.
........
  r57353 | georg.brandl | 2007-08-23 14:27:57 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1688564: document os.path.join's absolute path behavior in the docstring.
........
  r57354 | georg.brandl | 2007-08-23 14:36:05 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1625381: clarify match vs search introduction.
........
  r57355 | georg.brandl | 2007-08-23 14:42:54 -0700 (Thu, 23 Aug 2007) | 2 lines

  Bug #1758696: more info about descriptors.
........
  r57357 | georg.brandl | 2007-08-23 14:55:57 -0700 (Thu, 23 Aug 2007) | 2 lines

  Patch #1779550: remove redundant code in logging.
........
  r57378 | gregory.p.smith | 2007-08-23 22:11:38 -0700 (Thu, 23 Aug 2007) | 2 lines

  Fix bug 1725856.
........
  r57382 | georg.brandl | 2007-08-23 23:10:01 -0700 (Thu, 23 Aug 2007) | 2 lines

  uuid creation is now threadsafe, backport from py3k rev. 57375.
........
  r57389 | georg.brandl | 2007-08-24 04:47:37 -0700 (Fri, 24 Aug 2007) | 2 lines

  Bug #1765375: fix stripping of unwanted LDFLAGS.
........
  r57391 | guido.van.rossum | 2007-08-24 07:53:14 -0700 (Fri, 24 Aug 2007) | 2 lines

  Fix silly typo in test name.
........

2 files changed, 19 insertions, 9 deletions

diff --git a/Doc/reference/datamodel.rst b/Doc/reference/datamodel.rst
index 2f6013e..baa6eaa 100644
--- a/Doc/reference/datamodel.rst
+++ b/Doc/reference/datamodel.rst
@@ -1544,11 +1544,11 @@ Super Binding
    ``A.__dict__['m'].__get__(obj, A)``.
 
 For instance bindings, the precedence of descriptor invocation depends on the
-which descriptor methods are defined.  Data descriptors define both
-:meth:`__get__` and :meth:`__set__`.  Non-data descriptors have just the
+which descriptor methods are defined.  Normally, data descriptors define both
+:meth:`__get__` and :meth:`__set__`, while non-data descriptors have just the
 :meth:`__get__` method.  Data descriptors always override a redefinition in an
 instance dictionary.  In contrast, non-data descriptors can be overridden by
-instances.
+instances. [#]_
 
 Python methods (including :func:`staticmethod` and :func:`classmethod`) are
 implemented as non-data descriptors.  Accordingly, instances can redefine and
@@ -1817,6 +1817,9 @@ objects.  Immutable sequences methods should at most only define
 
    .. deprecated:: 2.0
       Support slice objects as parameters to the :meth:`__getitem__` method.
+      (However, built-in types in CPython currently still implement
+      :meth:`__getslice__`.  Therefore, you have to override it in derived
+      classes when implementing slicing.)
 
    Called to implement evaluation of ``self[i:j]``. The returned object should be
    of the same type as *self*.  Note that missing *i* or *j* in the slice
@@ -2112,6 +2115,13 @@ For more information on context managers, see :ref:`typecontextmanager`.
 .. [#] This, and other statements, are only roughly true for instances of new-style
    classes.
 
+.. [#] A descriptor can define any combination of :meth:`__get__`,
+   :meth:`__set__` and :meth:`__delete__`.  If it does not define :meth:`__get__`,
+   then accessing the attribute even on an instance will return the descriptor
+   object itself.  If the descriptor defines :meth:`__set__` and/or
+   :meth:`__delete__`, it is a data descriptor; if it defines neither, it is a
+   non-data descriptor.
+
 .. [#] For operands of the same type, it is assumed that if the non-reflected method
    (such as :meth:`__add__`) fails the operation is not supported, which is why the
    reflected method is not called.
diff --git a/Doc/reference/expressions.rst b/Doc/reference/expressions.rst
index 476c82d..ef71a80 100644
--- a/Doc/reference/expressions.rst
+++ b/Doc/reference/expressions.rst
@@ -741,7 +741,7 @@ less tightly than unary operators on its right.  The syntax is:
 
 Thus, in an unparenthesized sequence of power and unary operators, the operators
 are evaluated from right to left (this does not constrain the evaluation order
-for the operands).
+for the operands): ``-1**2`` results in ``-1``.
 
 The power operator has the same semantics as the built-in :func:`pow` function,
 when called with two arguments: it yields its left argument raised to the power
@@ -959,12 +959,12 @@ Comparisons can be chained arbitrarily, e.g., ``x < y <= z`` is equivalent to
 ``x < y and y <= z``, except that ``y`` is evaluated only once (but in both
 cases ``z`` is not evaluated at all when ``x < y`` is found to be false).
 
-Formally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *opa*, *opb*, ...,
-*opy* are comparison operators, then *a opa b opb c* ...*y opy z* is equivalent
-to *a opa b* :keyword:`and` *b opb c* :keyword:`and` ... *y opy z*, except that
-each expression is evaluated at most once.
+Formally, if *a*, *b*, *c*, ..., *y*, *z* are expressions and *op1*, *op2*, ...,
+*opN* are comparison operators, then ``a op1 b op2 c ... y opN z`` is equivalent
+to ``a op1 b and b op2 c and ... y opN z``, except that each expression is
+evaluated at most once.
 
-Note that *a opa b opb c* doesn't imply any kind of comparison between *a* and
+Note that ``a op1 b op2 c`` doesn't imply any kind of comparison between *a* and
 *c*, so that, e.g., ``x < y > z`` is perfectly legal (though perhaps not
 pretty).