Fred Drake [Sun, 28 Sep 2003 22:14:29 +0000 (22:14 +0000)]
- change computation of VERSION to use tools/getversioninfo; this is
more reliable than using the $Revision$ expansion
- $RELEASE is no longer needed; we can just use $VERSION now
Gregory P. Smith [Sat, 27 Sep 2003 23:00:19 +0000 (23:00 +0000)]
Use a threadsafe private DBEnv for each bsddb compatibility interface
db that is opened. DB_THREAD and DB_INIT_LOCK allow for multithreaded
access. DB_PRIVATE prevents the DBEnv from using the filesystem
(making it only usable by this process; and in this implementation
using one DBEnv per bsddb database)
* Fix markup.
* Fix entry order:
- >>> before ...
- __slots__ in the S section (like __future__ is in the F section)
Need to test the repaired(?) link to Guido's webpage.
Still needs to have the module reference links made relative to
the module directory instead of the tut directory. That will
require Fred's magic touch.
Fred Drake [Sat, 27 Sep 2003 16:04:23 +0000 (16:04 +0000)]
Make sure LaTeX2HTML's $TEXINPUTS variable is initialized to include
directories identified in the TEXINPUTS environment variable.
I think this is the last part of the fix for the version number
problems seen in the documentation for the 2.3.1 release.
Fred Drake [Sat, 27 Sep 2003 07:11:17 +0000 (07:11 +0000)]
Move content input files shared among the documents into a new directory
(commontex/), leaving only style support files in texinputs/. This makes
texinputs/ part of the formatting tools while commontex/ is strictly part
of the actual documentation.
Fred Drake [Sat, 27 Sep 2003 05:52:16 +0000 (05:52 +0000)]
Fix the most recent change to the invocation of the mkhowto script so
that it works for all targets.
The issue here is that there are two different levels in the directory
tree at which we execute mkhowto, so we can't define it just once
using a relative path (at least not with the current implementation
and Makefile structure). We use the GNUish $(shell) function here to
work around that restriction by identifying mkhowto using an absolute
path.
Tim Peters [Mon, 22 Sep 2003 18:38:53 +0000 (18:38 +0000)]
PlaySoundTest.test_alias_fallback(): Disabled this test, and explained
why in a new comment. My home Win98SE box is one of the "real systems"
alluded to (my system "default sound" appears to have vanished sometime
in the last month, that's certainly not a Python bug, and the MS
PlaySound docs are correct in their explanation of what happens then).
Bugfix candidate. If someone can still sneak it into 2.3.1, that would
be good.
Added test whether wchar_t is signed or not. A signed wchar_t is not usable as internal unicode type base for Py_UNICODE since the unicode implementation assumes an unsigned type.
Steve Purcell [Mon, 22 Sep 2003 11:08:12 +0000 (11:08 +0000)]
- Fixed loading of tests by name when name refers to unbound
method (PyUnit issue 563882, thanks to Alexandre Fayolle)
- Ignore non-callable attributes of classes when searching for test
method names (PyUnit issue 769338, thanks to Seth Falcon)
- New assertTrue and assertFalse aliases for comfort of JUnit users
- Automatically discover 'runTest()' test methods (PyUnit issue 469444,
thanks to Roeland Rengelink)
- Dropped Python 1.5.2 compatibility, merged appropriate shortcuts from
Python CVS; should work with Python >= 2.1.
- Removed all references to string module by using string methods instead
Gregory P. Smith [Sun, 21 Sep 2003 00:08:14 +0000 (00:08 +0000)]
Adds basic support for BerkeleyDB 4.2.x. Compiles and passes tests; new
features in BerkeleyDB not exposed. notably: the DB_MPOOLFILE interface
has not yet been wrapped in an object.
Adds support for building and installing bsddb3 in python2.3 that has
an older version of this module installed as bsddb without conflicts.
The pybsddb.sf.net build/packaged version of the module uses a
dynamicly loadable module called _pybsddb rather than _bsddb.
Tim Peters [Sat, 20 Sep 2003 22:16:26 +0000 (22:16 +0000)]
SF patch 809915: Fix bogus address to hopefully always break.
test_bad_address(): Recover from that VeriSign thought it would boost
its corporate coffers to start resolving http://www.sadflkjsasadf.com/.
Bugfix candidate -- although the bug is more VeriSign's than Python's!
Tim Peters [Sat, 20 Sep 2003 22:06:13 +0000 (22:06 +0000)]
test__locale (two underscores) can't pass on Windows: RADIXCHAR doesn't
exist, and neither do any of the specific 5-letter locale names the test
is looking for.
Jeremy Hylton [Tue, 16 Sep 2003 19:41:39 +0000 (19:41 +0000)]
Double-fix of crash in Unicode freelist handling.
If a length-1 Unicode string was in the freelist and it was
uninitialized or pointed to a very large (magnitude) negative number,
the check
unicode_latin1[unicode->str[0]] == unicode
could cause a segmentation violation, e.g. unicode->str[0] is 0xcbcbcbcb.
Fix this in two ways:
1. Change guard befor unicode_latin1[] to test against 256U. If I
understand correctly, the unsigned long used to store UCS4 on my
box was getting converted to a signed long to compare with the
signed constant 256.
2. Change _PyUnicode_New() to make sure the first element of str is
always initialized to zero. There are several places in the code
where the caller can exit with an error before initializing any
of str, which would leave junk in str[0].
Also, silence a compiler warning on pointer vs. int arithmetic.
Jeremy Hylton [Tue, 16 Sep 2003 03:41:45 +0000 (03:41 +0000)]
Change checks of PyUnicode_Resize() return value for clarity.
The unicode_resize() family only returns -1 or 0 so simply checking
for != 0 is sufficient, but somewhat unclear. Many Python API
functions return < 0 on error, reserving the right to return 0 or 1 on
success. Change the call sites for consistency with these calls.