Guido van Rossum [Fri, 22 Jun 2001 01:23:05 +0000 (01:23 +0000)]
Final update. Now Tim can cut the Windows installer and check in his
changes to PCbuild, then I'll tag the tree and build the final tar
file, and we can shove this puppy OUT OF THE DOOR...!
Guido van Rossum [Thu, 14 Jun 2001 11:26:45 +0000 (11:26 +0000)]
Fix two mentions of 2.1 instead of 2.0.1. Not worth fixing this in
the 2.0.1c1 release (which has already been built and pushed out to SF
and py.org), but will be correct in the 2.0.1 final release.
Does this require a matching update in the license as quoted in the
docs?
Barry Warsaw [Mon, 4 Jun 2001 19:35:04 +0000 (19:35 +0000)]
Backported two fixes from the Py2.1 tree (pre-unittest rewrite):
----------------------------
revision 1.3
date: 2001/04/10 15:01:20; author: gvanrossum; state: Exp; lines: +6 -0
Some other tests, when failing, don't always remove their TESTFN file.
Try to do it for them, so our mkdir() operation doesn't fail.
----------------------------
revision 1.2
date: 2001/03/02 05:46:17; author: gvanrossum; state: Exp; lines: +3 -3
When catching errors from os.rmdir(), test for os.error, not IOError!
----------------------------
Barry Warsaw [Mon, 4 Jun 2001 18:59:53 +0000 (18:59 +0000)]
The SF patch (#129288 - urllib.py - chanign %02x to %02X in quoting)
was applied to urllib.py, but the corresponding change to
test_urllib.py was not applied. Backport revision 1.6 of this file
into the 2.0 maintenance branch.
----------------------------
revision 1.6
date: 2001/01/19 07:00:08; author: tim_one; state: Exp; lines: +8 -3
urllib.py very recently changed to produce uppercase escapes, but no
corresponding changes were made to its std test.
----------------------------
Inspired by Tim Peters' 1.14->1.15 checkin to asynchat
Quoting orignal message:
'''
Fix from the Madusa mailing list:
http://groups.yahoo.com/group/medusa/message/333
It's clear that Medusa should not be checking for an empty buffer
via "buf is ''". The patch merely changes "is" to "==". However,
there's a mystery here all the same: Python attempts to store null
strings uniquely, so it's unclear why "buf is ''" ever returned
false when buf actually was empty. *Some* string operations produce
non-unique null strings, e.g.
'''
Checking the diff from PC/getpatchp.c 1.22 to 1.23
This fixes many bugs, and was proposed by Thomas Heller.
Original log message:
'''
Checkin updated version of patch #103933 . As Thomas says, fixes the bugs
#131064, #129584, #127722. See the discussion in bug #131064
'''
Moshe Zadka [Sat, 31 Mar 2001 16:09:32 +0000 (16:09 +0000)]
- #233253 - distutils/command/build_ext.py - the --define and --undef options
didn't work, whether specified on the command-line or in setup.cfg.
- distutils/command/build_ext.py - make docstrings raw
- #128930 - distutils/command/build_ext.py - split rpath argument
Suggested by AMK, but had to be massaged a bit from the cvs diff
Moshe Zadka [Sat, 31 Mar 2001 14:58:20 +0000 (14:58 +0000)]
- #227562 - urllib.py - call URLopener.http_error_default when
an invalid 401 request is being handled.
- urllib.py - provide simple recovery/escape from apparent redirect recursion
- #129288 - urllib.py - chanign %02x to %02X in quoting
- urllib.py - HTTPS now works with string URLs
Moshe Zadka [Sat, 31 Mar 2001 14:26:54 +0000 (14:26 +0000)]
- xmllib.py - Moved clearing of "literal" flag. The flag is set in setliteral
which can be called from a start tag handler. When the corresponding end
tag is read the flag is cleared. However, it didn't get cleared when
the start tag was for an empty element of the type <tag .../>. This
modification fixes the problem.
Moshe Zadka [Fri, 30 Mar 2001 21:01:09 +0000 (21:01 +0000)]
- exceptions.c - make_class() Added a "goto finally" so that if
populate_methods() fails, the return status will be -1 (failure)
instead of 0 (success).
fini_exceptions(): When decref'ing the static pointers to the
exception classes, clear out their dictionaries too. This breaks a
cycle from class->dict->method->class and allows the classes with
unbound methods to be reclaimed. This plugs a large memory leak in a
common Py_Initialize()/dosomething/Py_Finalize() loop.
Moshe Zadka [Fri, 30 Mar 2001 17:20:58 +0000 (17:20 +0000)]
#126161 and 123634 -- Unicode strings could not be pickled correctly.
This is *backwards incompatible* with the previous pickling scheme,
which wasnot reversible
Moshe Zadka [Fri, 30 Mar 2001 16:31:50 +0000 (16:31 +0000)]
SF bug 119622: compile errors due to redundant atof decls.
Fixed in compile.c and marshal.c
Quote tim: "Python shouldn't be in the business of defining atof"
Moshe Zadka [Thu, 29 Mar 2001 19:27:29 +0000 (19:27 +0000)]
Yay! First fix I'm commiting on the way to 2.0.1
This is supposed to fix compilation problems of the curses
module on SGI, and is based on the patch from 2.39 to 2.40
This should fix bug #116172
From the 2.39->2.40 log:
* Check for 'sgi' preprocessor symbol, not '__sgi__'
* Surround individual character macros with #ifdef's, instead of making them
all rely on STRICT_SYSV_CURSES
Metacomment: this is as much to test procedures as it is
to actually fix the bug. If you're a CVS wizard please look
at the checkin message to see I'm commiting correctly,
and e-mail me if not, and I'll do my best to back it out and
reread the docos.
Tim Peters [Mon, 16 Oct 2000 17:35:13 +0000 (17:35 +0000)]
Test for math.* exceptional behavior only in verbose mode, so that the
oddball platforms (where, e.g., math.exp(+huge) still fails to raise
OverflowError) don't fail the std test suite when run normally.
Barry Warsaw [Mon, 16 Oct 2000 15:47:50 +0000 (15:47 +0000)]
find(): Application of (slightly modified) SF patch #101928 by Ulf
Betlehem, verified by Peter Funk. Fixes preservation of language
search order lost due to use of dictionary keys instead of a list.
Closes SF bug #116964.