Walter Dörwald [Tue, 21 Dec 2004 22:35:24 +0000 (22:35 +0000)]
Backport checkin:
The changes to the stateful codecs in 2.4 resulted in StreamReader.readline()
trying to return a complete line even if a size parameter was given (see
http://www.python.org/sf/1076985). This leads to buffer overflows with long
source lines under Windows if e.g. cp1252 is used as the source encoding.
This patch reverts the behaviour of readline() to something that behaves more
like Python 2.3: If a size parameter is given, read() is called only once.
As a side effect of this, readline() now supports all types of linebreaks
supported by unicode.splitlines().
Note that the tokenizer is still broken and it's possible to provoke segfaults
(see http://www.python.org/sf/1089395).
Gregory P. Smith [Sun, 19 Dec 2004 22:25:32 +0000 (22:25 +0000)]
* reworked berkeleydb include and library detection code in setup.py
to make sure the library version that matches the header file is used.
* update the bsddb module docs to indicate BerkeleyDB 3.2 thru 4.3 support
Fred Drake [Sun, 5 Dec 2004 07:45:17 +0000 (07:45 +0000)]
once we're post-release, we really need to bump the version number;
the documentation build gets the version number from here as well, and
the script that updates www.python.org/dev/doc/ needs to get an
updated version number
Barry Warsaw [Sun, 5 Dec 2004 03:34:14 +0000 (03:34 +0000)]
Fixes for SF #1076485, which I'll apply to the CVS head too. The problem was
caused by a self._input.readline() call that wasn't checking for the
NeedsMoreData marker.
msg_43.txt contains a message that illustrates the problem, when
email.message_from_*() is called. That interface uses the Parser API, which
splits reads into 8192 byte chunks. It so happens that for the test message,
the 8192 chunk falls inside a message/delivery-status, which is where in the
FeedParser the readline() call was that didn't check for NeedsMoreData.
I also added an assert to unreadline() so it'll be more evident if an attempt
to push back NeedsMoreData ever happens again.
Fred Drake [Wed, 1 Dec 2004 06:29:51 +0000 (06:29 +0000)]
- fix up internal hyperlink generation in PDF formatting so that links at
the beginning of a paragraph do not generate errors; this affected
things like \refmodule when it came first in a paragraph
- clean up the .sty file to separate out the treatment of the start
of a new paragraph
Barry Warsaw [Sun, 28 Nov 2004 00:21:42 +0000 (00:21 +0000)]
Fix for SF bug #1072623. When the last line of the input string does not end
in a newline, and it's an end boundary, the FeedParser wasn't recognizing it
as such. Tweak the regexp to make the ending linesep optional.
For grins, clear self._partial when closing the BufferedSubFile.
Hye-Shik Chang [Thu, 25 Nov 2004 04:04:20 +0000 (04:04 +0000)]
Rename a static variable "history_length" to "_history_length".
GNU readline exports a global variable that has such a name already
and the collision makes gcc4 doesn't compile the source.
Brett Cannon [Wed, 24 Nov 2004 03:01:36 +0000 (03:01 +0000)]
Have testLoadTkFailure() skip on cygwin since Tcl/Tk on cygwin renders to the
Windows GDI directly and thus does not need a DISPLAY environment variable.
Kurt B. Kaiser [Tue, 23 Nov 2004 18:06:08 +0000 (18:06 +0000)]
Hye-Shik Chang's fix for Bug 875692.
Improve signal handling, especially when using threads, by forcing an early
re-execution of PyEval_EvalFrame() "periodic" code when things_to_do is not
cleared by Py_MakePendingCalls().
Tim Peters [Mon, 22 Nov 2004 16:49:02 +0000 (16:49 +0000)]
SF bug 1071087: os.walk example for deleting a full tree is sometime wrong.
Clarify that the example assumes no links are present; the point of the
example is to illustrate a need for topdown=False, not to wrestle with
platform-dependent link convolutions.
Also spell os.path.join() out in full, instead of using a shortcut import.
The bug reporter was confused by that too, and it's clearer this way.
Bugfix candidate; but I don't intend to backport it.
Removed a section that is out of date after type/class unification.
While there was still some validity, the paragraph offered more
confusion that insight.
Skip Montanaro [Wed, 17 Nov 2004 16:04:15 +0000 (16:04 +0000)]
split functionality into pystack and pystackv commands. The former will
work with core dumps because it avoids calling any Python API routines. The
latter prints all the local variable values as well as the stack frames but
won't work with core dumps because it relies on _PyObject_Dump to print
variables.
Kurt B. Kaiser [Tue, 16 Nov 2004 21:31:08 +0000 (21:31 +0000)]
Saving a Keyset w/o making changes (by using the "Save as New Custom Key Set"
button) caused IDLE to fail on restart (no new keyset was created in
config-keys.cfg). Also true for Theme/highlights. Python Bug 1064535.