]> granicus.if.org Git - python/log
python
20 years agoFixed doctest error (wrong prompts)
Edward Loper [Mon, 9 Aug 2004 02:03:30 +0000 (02:03 +0000)]
Fixed doctest error (wrong prompts)

20 years agoPrevious commit was viewed as "perverse". Changed to just cast the unused
Brett Cannon [Sun, 8 Aug 2004 21:21:18 +0000 (21:21 +0000)]
Previous commit was viewed as "perverse".  Changed to just cast the unused
variable to void..

Thanks to Sjoerd Mullender for the suggested change.

20 years agoAdd a test for Context.copy().
Raymond Hettinger [Sun, 8 Aug 2004 20:17:45 +0000 (20:17 +0000)]
Add a test for Context.copy().

20 years agoTell unittest that source files with "badsyntax" in their names should
Tim Peters [Sun, 8 Aug 2004 16:43:59 +0000 (16:43 +0000)]
Tell unittest that source files with "badsyntax" in their names should
raise SyntaxError.  test_compiler passes now on WinXP, at least in a
release-build non-O run.

20 years agoRenamed locals to better reflect their meanings.
Tim Peters [Sun, 8 Aug 2004 16:37:37 +0000 (16:37 +0000)]
Renamed locals to better reflect their meanings.

20 years agoIn verbose mode, display the name of each file before trying to compile
Tim Peters [Sun, 8 Aug 2004 16:32:54 +0000 (16:32 +0000)]
In verbose mode, display the name of each file before trying to compile
it.  Else when this fails, there's no way to tell which file it was
chewing on.

20 years agoBrief xrange news.
Tim Peters [Sun, 8 Aug 2004 07:24:22 +0000 (07:24 +0000)]
Brief xrange news.

20 years agoBug 1003935: xrange overflows
Tim Peters [Sun, 8 Aug 2004 07:17:39 +0000 (07:17 +0000)]
Bug 1003935:  xrange overflows

Added XXX comment about why the undocumented PyRange_New() API function
is too broken to be worth the considerable pain of repairing.

Changed range_new() to stop using PyRange_New().  This fixes a variety
of bogus errors.  Nothing in the core uses PyRange_New() now.

Documented that xrange() is intended to be simple and fast, and that
CPython restricts its arguments, and length of its result sequence, to
native C longs.

Added some tests that failed before the patch, and repaired a test that
relied on a bogus OverflowError getting raised.

20 years agoTrimmed trailing whitespace.
Tim Peters [Sun, 8 Aug 2004 06:29:10 +0000 (06:29 +0000)]
Trimmed trailing whitespace.

20 years agoGet rid of the ignore_imports argument to DocTestFinder.find().
Tim Peters [Sun, 8 Aug 2004 06:11:48 +0000 (06:11 +0000)]
Get rid of the ignore_imports argument to DocTestFinder.find().
This got slammed in when find() was fixed to stop grabbing doctests
from modules imported *by* the module being tested.  Such tests cannot
be expected to succeed, since they'll be run with the current module's
globals.  Dozens of Zope3 doctests were failing because of that.

It wasn't clear why ignore_imports got added then.  Maybe it's because
some existing tests failed when the change was made.  Whatever, it's
a Bad Idea so it's gone now.

The only use of it was exceedingly obscure, in test_doctest's "Duplicate
Removal" test.  It was "needed" there because, as an artifact of running
a doctest inside a doctest, the func_globals of functions compiled in
the second-level doctest don't match the module globals, and so the
test-finder believed these functions were from a foreign module and
skipped them.  But that took a long time to figure out, and I actually
understand some of this stuff <0.9 wink>.

That problem was resolved by moving the source code for the second-level
doctest into an actual module (test/doctest_aliases.py).

The only remaining difficulty was that the test for the deprecated
Tester.rundict() then failed, because the test finder doesn't take
module=None at face value, trying to guess which module the user really
intended then.  Its guess wasn't appropriate for what Tester.rundict
needs when module=None is given to *it*, which is "no, there is no
module here, and I mean it".  So now passing module=False means exactly
that.  This is hokey, but ignore_imports=False was really a hack to worm
around that there was no way to tell the test-finder that module=None
*sometimes* means what it says.  There was no use case for the combination
of passing a real module with ignore_imports=False.

20 years ago* Context.copy() now makes a deepcopy.
Raymond Hettinger [Sun, 8 Aug 2004 04:03:24 +0000 (04:03 +0000)]
* Context.copy() now makes a deepcopy.
* Facilitate reloads of local thread.

20 years agoType in docstring.
Tim Peters [Sun, 8 Aug 2004 03:38:33 +0000 (03:38 +0000)]
Type in docstring.

20 years agoAlso deprecated the old Tester class, which is no longer used by anything
Tim Peters [Sun, 8 Aug 2004 02:43:33 +0000 (02:43 +0000)]
Also deprecated the old Tester class, which is no longer used by anything
except internal tests.

20 years agoDeprecate the doctest.is_private() function.
Tim Peters [Sun, 8 Aug 2004 01:52:57 +0000 (01:52 +0000)]
Deprecate the doctest.is_private() function.

20 years agoDeprecated testmod's useless & confusing isprivate gimmick.
Tim Peters [Sun, 8 Aug 2004 01:48:59 +0000 (01:48 +0000)]
Deprecated testmod's useless & confusing isprivate gimmick.
Ripped out the docs for the new DocTestFinder's namefilter argument,
and renamed it to _namefilter; this only existed to support isprivate.
Removed the new DocTestFinder's objfilter argument.  No point adding
more cruft to a broken filtering design.

20 years agoWhitespace normalization.
Tim Peters [Sun, 8 Aug 2004 01:05:14 +0000 (01:05 +0000)]
Whitespace normalization.

20 years agoBug 1003471: Python 1.5.2 security vulnerability still present in 2.3.4
Tim Peters [Sun, 8 Aug 2004 01:00:47 +0000 (01:00 +0000)]
Bug 1003471: Python 1.5.2 security vulnerability still present in 2.3.4

That's the title of the report, but the hole was probably plugged since
Python 2.0.  See corresponding checkin to PC/getpathp.c:  a crucial
precondition for joinpath() was neither documented nor verified, and there
are so many callers with so many conditional paths that no "eyeball
analysis" is satisfactory.  Now Python dies with a fatal error if the
precondition isn't satisfied, instead of allowing a buffer overrun.

NOT TESTED!  The Windows version of the patch was, but not this one.  I
don't feel like waiting for someone to notice the patch I attached to the
bug report.  If it doesn't compile, sorry, but fix it <wink>.  If it
does compile, it's "obviously correct".

20 years agoPurged Berkeley code of references to the long-obsolete whrandom.
Tim Peters [Sun, 8 Aug 2004 00:54:21 +0000 (00:54 +0000)]
Purged Berkeley code of references to the long-obsolete whrandom.

20 years agoTrigger DeprecationWarning
Andrew M. Kuchling [Sat, 7 Aug 2004 21:44:37 +0000 (21:44 +0000)]
Trigger DeprecationWarning

20 years agoRemove MacOS 9 ref
Andrew M. Kuchling [Sat, 7 Aug 2004 21:35:06 +0000 (21:35 +0000)]
Remove MacOS 9 ref

20 years agoRemove MacOS 9 refs
Andrew M. Kuchling [Sat, 7 Aug 2004 21:33:44 +0000 (21:33 +0000)]
Remove MacOS 9 refs

20 years ago[Patch #862531] Update version numbers.
Andrew M. Kuchling [Sat, 7 Aug 2004 21:30:13 +0000 (21:30 +0000)]
[Patch #862531] Update version numbers.

20 years agoLet's not use string exceptions any more.
Armin Rigo [Sat, 7 Aug 2004 21:27:43 +0000 (21:27 +0000)]
Let's not use string exceptions any more.

20 years agoFix
Michael W. Hudson [Sat, 7 Aug 2004 21:13:46 +0000 (21:13 +0000)]
Fix

[ 777659 ] Uninitialized variable used in Tools/faqwiz/faqwiz.py

with help from jlgijsbers on #python-dev IRC.

20 years agoThis was quite a dark bug in my recent in-place string concatenation
Armin Rigo [Sat, 7 Aug 2004 20:58:32 +0000 (20:58 +0000)]
This was quite a dark bug in my recent in-place string concatenation
hack: it would resize *interned* strings in-place!  This occurred because
their reference counts do not have their expected value -- stringobject.c
hacks them.  Mea culpa.

20 years agoRemoving tests that fail because of changes in PyString_InternInPlace(),
Armin Rigo [Sat, 7 Aug 2004 20:30:03 +0000 (20:30 +0000)]
Removing tests that fail because of changes in PyString_InternInPlace(),
as discussed on IRC.  The equivalent tests for the new behavior are in
test_builtin.py.

20 years ago[Patch #999280 ] Update kwargs in pickle docs to match implementations
Andrew M. Kuchling [Sat, 7 Aug 2004 20:25:55 +0000 (20:25 +0000)]
[Patch #999280 ] Update kwargs in pickle docs to match implementations

20 years ago[Patch #1003861 from Dima Dorfman] Fix markup in concrete.tex:
Andrew M. Kuchling [Sat, 7 Aug 2004 20:19:24 +0000 (20:19 +0000)]
[Patch #1003861 from Dima Dorfman] Fix markup in concrete.tex:
PyObject* o -> PyObject *o to be consistent with the
rest of the file
- Correct markup for Py_True
- Remove duplicate description of PyBool_Check

20 years agoUse LaTeX markup
Andrew M. Kuchling [Sat, 7 Aug 2004 20:17:48 +0000 (20:17 +0000)]
Use LaTeX markup

20 years ago[Bug #914375] Crude modulefinder docs, but the module's code is very hard to read...
Andrew M. Kuchling [Sat, 7 Aug 2004 20:13:34 +0000 (20:13 +0000)]
[Bug #914375] Crude modulefinder docs, but the module's code is very hard to read and it's not clear which bits are public and which aren't.  The module's author should really be doing this (and using docstrings in the code, too)

20 years agoeval_frame now has a PyAPI-style name
Michael W. Hudson [Sat, 7 Aug 2004 20:11:22 +0000 (20:11 +0000)]
eval_frame now has a PyAPI-style name

20 years agoFixed some compiler warnings.
Armin Rigo [Sat, 7 Aug 2004 19:27:39 +0000 (19:27 +0000)]
Fixed some compiler warnings.

20 years agoAdd a trivial test for the compiler package, guarded by compiler resource.
Jeremy Hylton [Sat, 7 Aug 2004 19:25:33 +0000 (19:25 +0000)]
Add a trivial test for the compiler package, guarded by compiler resource.

This test is insanely slow, so it requires a resource.  On my machine,
it also appears to dump core.  I think the problem is a stack
overflow, but haven't been able to confirm.

20 years agoRemove various modules that have been documented
Andrew M. Kuchling [Sat, 7 Aug 2004 19:21:59 +0000 (19:21 +0000)]
Remove various modules that have been documented

20 years agoSF patch 836879.
Jeremy Hylton [Sat, 7 Aug 2004 19:21:56 +0000 (19:21 +0000)]
SF patch 836879.

Don't generate code for asserts in -O mode.

20 years agoSubclasses of string can no longer be interned. The semantics of
Jeremy Hylton [Sat, 7 Aug 2004 19:20:05 +0000 (19:20 +0000)]
Subclasses of string can no longer be interned.  The semantics of
interning were not clear here -- a subclass could be mutable, for
example -- and had bugs.  Explicitly interning a subclass of string
via intern() will raise a TypeError.  Internal operations that attempt
to intern a string subclass will have no effect.

Added a few tests to test_builtin that includes the old buggy code and
verifies that calls like PyObject_SetAttr() don't fail.  Perhaps these
tests should have gone in test_string.

20 years ago[Bug #827209] Add footnote from Alex M. about listcomps leaking their index variables
Andrew M. Kuchling [Sat, 7 Aug 2004 19:16:32 +0000 (19:16 +0000)]
[Bug #827209] Add footnote from Alex M. about listcomps leaking their index variables

20 years agoSF bug 1003471: Python 1.5.2 security vulnerability
Tim Peters [Sat, 7 Aug 2004 19:12:27 +0000 (19:12 +0000)]
SF bug 1003471:  Python 1.5.2 security vulnerability

This was probably fixed in rev 1.32 of getpath.c, but there are so
many paths thru the code that invoke joinpath() it's not at all
obvious that it *is* fixed.  It doesn't help confidence that a crucial
precondition for calling joinpath() was neither documented nor verified.
It is now, and joinpath() will barf with a fatal error now rather than
overrun the buffer, if the precondition isn't met.

Note that this patch only changes the Windows flavor.  I attached another
patch to the bug report for the POSIX flavor (which I can't test
conveniently).

20 years agoVarious minor edits
Andrew M. Kuchling [Sat, 7 Aug 2004 19:10:36 +0000 (19:10 +0000)]
Various minor edits

20 years ago[Bug #866222] Update docs to match the module
Andrew M. Kuchling [Sat, 7 Aug 2004 19:06:48 +0000 (19:06 +0000)]
[Bug #866222] Update docs to match the module

20 years ago[Bug #866222] Update docstrings.
Andrew M. Kuchling [Sat, 7 Aug 2004 19:02:19 +0000 (19:02 +0000)]
[Bug #866222] Update docstrings.

20 years agoFix
Michael W. Hudson [Sat, 7 Aug 2004 17:57:16 +0000 (17:57 +0000)]
Fix

[ 991812 ] PyArg_ParseTuple can miss errors with warnings as exceptions

as suggested in the report.

This is definitely a 2.3 candidate (as are most of the checkins I've
made in the last month...)

20 years ago[Bug #991883] Document calling condition
Andrew M. Kuchling [Sat, 7 Aug 2004 17:53:05 +0000 (17:53 +0000)]
[Bug #991883] Document calling condition

20 years ago[Bug #990792] Mention that repl can be a callable
Andrew M. Kuchling [Sat, 7 Aug 2004 17:41:54 +0000 (17:41 +0000)]
[Bug #990792] Mention that repl can be a callable

20 years agoFix urllib2.urlopen() handling of chunked content encoding.
Jeremy Hylton [Sat, 7 Aug 2004 17:40:50 +0000 (17:40 +0000)]
Fix urllib2.urlopen() handling of chunked content encoding.

The change to use the newer httplib interface admitted the possibility
that we'd get an HTTP/1.1 chunked response, but the code didn't handle
it correctly.  The raw socket object can't be pass to addinfourl(),
because it would read the undecoded response.  Instead, addinfourl()
must call HTTPResponse.read(), which will handle the decoding.

One extra wrinkle is that the HTTPReponse object can't be passed to
addinfourl() either, because it doesn't implement readline() or
readlines().  As a quick hack, use socket._fileobject(), which
implements those methods on top of a read buffer.  (suggested by mwh)

Finally, add some tests based on test_urllibnet.

Thanks to Andrew Sawyers for originally reporting the chunked problem.

20 years agoFix
Michael W. Hudson [Sat, 7 Aug 2004 17:39:35 +0000 (17:39 +0000)]
Fix

1000841 ] "make pdf" failure w/ 2.4 docs

in the suggested way, by uglifying a URL.

20 years ago[Bug #998066] Mention result mismatch
Andrew M. Kuchling [Sat, 7 Aug 2004 17:28:17 +0000 (17:28 +0000)]
[Bug #998066] Mention result mismatch

20 years ago[Bug #923315] Produce correct result on AIX
Andrew M. Kuchling [Sat, 7 Aug 2004 17:21:27 +0000 (17:21 +0000)]
[Bug #923315] Produce correct result on AIX

20 years agoPatch:
Michael W. Hudson [Sat, 7 Aug 2004 17:05:42 +0000 (17:05 +0000)]
Patch:

1003863 ] adds an index entry for __all__ to the tutorial

20 years ago[Bug #873146] Document pickletools module (haven't tested the LaTeX yet)
Andrew M. Kuchling [Sat, 7 Aug 2004 16:53:59 +0000 (16:53 +0000)]
[Bug #873146] Document pickletools module (haven't tested the LaTeX yet)

20 years agoAdd argument to docstring
Andrew M. Kuchling [Sat, 7 Aug 2004 16:51:30 +0000 (16:51 +0000)]
Add argument to docstring

20 years agoPoint out that the setdefault defaults the value to None. Inspired by
Michael W. Hudson [Sat, 7 Aug 2004 16:41:34 +0000 (16:41 +0000)]
Point out that the setdefault defaults the value to None.  Inspired by
Michael Chermside's thinking about patch #748126 (the chief upshot of
which thinking was "reject it!").

20 years agoResolution of bug #997368, "strftime() backward compatibility".
Barry Warsaw [Sat, 7 Aug 2004 16:38:40 +0000 (16:38 +0000)]
Resolution of bug #997368, "strftime() backward compatibility".

Specifically, time.strftime() no longer accepts a 0 in the yday position of a
time tuple, since that can crash some platform strftime() implementations.

parsedate_tz(): Change the return value to return 1 in the yday position.

Update tests in test_rfc822.py and test_email.py

20 years agoSF bug 874842 and patch 997626: httplib bugs
Jeremy Hylton [Sat, 7 Aug 2004 16:28:14 +0000 (16:28 +0000)]
SF bug 874842 and patch 997626: httplib bugs

Hack httplib to work with broken Akamai proxies.
Make sure that httplib doesn't add extract Accept-Encoding or
Content-Length headers if the client has already set them.

20 years agoMake 'bin' argument trigger DeprecationWarning
Andrew M. Kuchling [Sat, 7 Aug 2004 16:27:24 +0000 (16:27 +0000)]
Make 'bin' argument trigger DeprecationWarning

20 years ago[Bug #984952] Include more material from PEP 307.
Andrew M. Kuchling [Sat, 7 Aug 2004 16:24:18 +0000 (16:24 +0000)]
[Bug #984952] Include more material from PEP 307.
I haven't tried to include all the material on old-style classes using protocols 0,1.  The details are lengthy; someone who knows
more about the pickle module should decide if they're important enough
to be in the docs or not.

20 years agoResolution of SF bug #1002475 and patch #1003693; Header lines that end in
Barry Warsaw [Sat, 7 Aug 2004 15:57:52 +0000 (15:57 +0000)]
Resolution of SF bug #1002475 and patch #1003693; Header lines that end in
\r\n only get the \n stripped, not the \r (unless it's the last header which
does get the \r stripped).  Patch by Tony Meyer.

test_whitespace_continuation_last_header(),
test_strip_line_feed_and_carriage_return_in_headers(): New tests.

_parse_headers(): Be sure to strip \r\n from the right side of header lines.

20 years ago[Bug #984952] Include some material from PEP 307
Andrew M. Kuchling [Sat, 7 Aug 2004 15:49:24 +0000 (15:49 +0000)]
[Bug #984952] Include some material from PEP 307

20 years agoThis is more-or-less:
Michael W. Hudson [Sat, 7 Aug 2004 15:27:16 +0000 (15:27 +0000)]
This is more-or-less:

1005123 ] test_curses fails on MacOS X 10.3

Be a bit more guarded about what we expect a terminal to be capable
of.

20 years agoSomehow (no idea how!) I missed half of patch #1005008. Sorry about that.
Michael W. Hudson [Sat, 7 Aug 2004 15:20:15 +0000 (15:20 +0000)]
Somehow (no idea how!) I missed half of patch #1005008.   Sorry about that.

20 years agoThis is patch
Michael W. Hudson [Sat, 7 Aug 2004 15:18:07 +0000 (15:18 +0000)]
This is patch

1005008 ] curses.wrapper should also forward keyword args

Plus my rewrite to use finally as opposed to painfully doing the
equivalent by hand.

20 years ago[Bug #998307] Use open() instead of file() in docs
Andrew M. Kuchling [Sat, 7 Aug 2004 15:11:24 +0000 (15:11 +0000)]
[Bug #998307] Use open() instead of file() in docs

20 years agofix two typos in markup
Fred Drake [Sat, 7 Aug 2004 14:28:37 +0000 (14:28 +0000)]
fix two typos in markup

20 years agoAdd name
Andrew M. Kuchling [Sat, 7 Aug 2004 14:17:50 +0000 (14:17 +0000)]
Add name

20 years agoCreate section for 2.4a3; remove empty sections in 2.4a2
Andrew M. Kuchling [Sat, 7 Aug 2004 14:03:33 +0000 (14:03 +0000)]
Create section for 2.4a3; remove empty sections in 2.4a2

20 years agoAnother typo
Andrew M. Kuchling [Sat, 7 Aug 2004 14:00:39 +0000 (14:00 +0000)]
Another typo

20 years agoTypo fix
Andrew M. Kuchling [Sat, 7 Aug 2004 13:59:22 +0000 (13:59 +0000)]
Typo fix

20 years agoAdd recent items
Andrew M. Kuchling [Sat, 7 Aug 2004 13:58:02 +0000 (13:58 +0000)]
Add recent items

20 years agoSimplify language
Andrew M. Kuchling [Sat, 7 Aug 2004 13:24:12 +0000 (13:24 +0000)]
Simplify language

20 years agoAdd string concat item
Andrew M. Kuchling [Sat, 7 Aug 2004 13:13:31 +0000 (13:13 +0000)]
Add string concat item

20 years agoExercise DocTestSuite's search for __test__.
Raymond Hettinger [Sat, 7 Aug 2004 06:15:12 +0000 (06:15 +0000)]
Exercise DocTestSuite's search for __test__.

20 years agoWhitespace normalization.
Tim Peters [Sat, 7 Aug 2004 06:03:09 +0000 (06:03 +0000)]
Whitespace normalization.

20 years agoBug 772091: doctest.DocTestSuite does not support __test__
Tim Peters [Sat, 7 Aug 2004 05:37:52 +0000 (05:37 +0000)]
Bug 772091:  doctest.DocTestSuite does not support __test__

This got fixed "by magic" as part of the refactoring, but wasn't tested
as such.  Now it is.

20 years agoSF bug #1004669: Type returned from .keys() is not checked
Raymond Hettinger [Sat, 7 Aug 2004 04:55:30 +0000 (04:55 +0000)]
SF bug #1004669:  Type returned from .keys() is not checked

20 years agoSF bug #1002530: test_decimal fails if repeated
Raymond Hettinger [Fri, 6 Aug 2004 23:42:16 +0000 (23:42 +0000)]
SF bug #1002530:  test_decimal fails if repeated

* Protect the pre-defined contexts by using a deepcopy() instead of copy().
* Micro-optimization:  prefer x&1 over x%2

20 years agoMerging from tim-doctest-branch, which is now closed.
Tim Peters [Fri, 6 Aug 2004 22:02:59 +0000 (22:02 +0000)]
Merging from tim-doctest-branch, which is now closed.
This primarily adds more powerful ways to work with unittest, including
spiffy support for building suites out of doctests in non-Python
"text files".

20 years agoSF bug #1004088: big code objects (>64K) may be optimized incorrectly
Raymond Hettinger [Fri, 6 Aug 2004 19:46:34 +0000 (19:46 +0000)]
SF bug #1004088:   big code objects (>64K) may be optimized incorrectly
Will backport.

20 years agoBump version
Andrew M. Kuchling [Fri, 6 Aug 2004 18:55:48 +0000 (18:55 +0000)]
Bump version

20 years agoTypo fix; use 'in-place' instead of 'inplace'
Andrew M. Kuchling [Fri, 6 Aug 2004 18:55:09 +0000 (18:55 +0000)]
Typo fix; use 'in-place' instead of 'inplace'

20 years agoSF patch #980695: efficient string concatenation
Raymond Hettinger [Fri, 6 Aug 2004 18:43:09 +0000 (18:43 +0000)]
SF patch #980695:  efficient string concatenation
(Original patch by Armin Rigo).

20 years agoChanges the remote address used for tests in TimeoutTestCase from google.com to
Brett Cannon [Fri, 6 Aug 2004 04:30:46 +0000 (04:30 +0000)]
Changes the remote address used for tests in TimeoutTestCase from google.com to
python.org .  This way the delay should be great enough for
testConnectTimeout() to pass even when one has a really fast Net connection
that allows connections faster than .001 seconds.

20 years agofix markup nit, typo
Fred Drake [Fri, 6 Aug 2004 03:34:20 +0000 (03:34 +0000)]
fix markup nit, typo

20 years agoupdate to reflect the new significance of the "@" character
Fred Drake [Thu, 5 Aug 2004 21:11:27 +0000 (21:11 +0000)]
update to reflect the new significance of the "@" character

20 years agoAdded new codecs and aliases for ISO_8859-11, ISO_8859-16 and
Marc-André Lemburg [Thu, 5 Aug 2004 12:43:30 +0000 (12:43 +0000)]
Added new codecs and aliases for ISO_8859-11, ISO_8859-16 and
TIS-620.

Closes SF bug #1001895: Adding missing ISO 8859 codecs, especially Thai.

20 years agoWarn that settrace() is implementation rather than language
Phillip J. Eby [Thu, 5 Aug 2004 12:13:46 +0000 (12:13 +0000)]
Warn that settrace() is implementation rather than language
definition, per Guido's request  in off-list email.

20 years agoThis commit was manufactured by cvs2svn to create tag 'r24a2'. v2.4a2
cvs2svn [Thu, 5 Aug 2004 07:21:01 +0000 (07:21 +0000)]
This commit was manufactured by cvs2svn to create tag 'r24a2'.

20 years agorelease date
Anthony Baxter [Thu, 5 Aug 2004 07:21:01 +0000 (07:21 +0000)]
release date

20 years agoDoc/lib/libfuncs.tex: correct items in itemize
Matthias Klose [Wed, 4 Aug 2004 23:18:49 +0000 (23:18 +0000)]
Doc/lib/libfuncs.tex: correct items in itemize
Doc/dist/dist.tex: avoid uncommenting the verbose environment (by the py2texi tool)

20 years agoadd constants for many error values added over the past couple of
Fred Drake [Wed, 4 Aug 2004 22:28:16 +0000 (22:28 +0000)]
add constants for many error values added over the past couple of
years

20 years agoExample.__init__: this cannot use assert, because that fails to trigger
Tim Peters [Wed, 4 Aug 2004 20:04:32 +0000 (20:04 +0000)]
Example.__init__:  this cannot use assert, because that fails to trigger
in a -O run, and so test_doctest was failing under -O.  Simple cause,
simple cure.

20 years agoEdward Loper's cool and massive refactoring of doctest.py, merged from
Tim Peters [Wed, 4 Aug 2004 18:46:34 +0000 (18:46 +0000)]
Edward Loper's cool and massive refactoring of doctest.py, merged from
the tim-doctest-merge-24a2 tag on the the tim-doctest-branch branch.
We did development on the branch in case it wouldn't land in time for
2.4a2, but the branch looked good:  Edward's tests passed there, ditto
Python's tests, and ditto the Zope3 tests.  Together, those hit doctest
heavily.

20 years agoAdd a proper pointer to SF bug item.
Hye-Shik Chang [Wed, 4 Aug 2004 17:40:38 +0000 (17:40 +0000)]
Add a proper pointer to SF bug item.

20 years agoSF #941229: Decode source code with sys.stdin.encoding in interactive
Hye-Shik Chang [Wed, 4 Aug 2004 17:36:41 +0000 (17:36 +0000)]
SF #941229: Decode source code with sys.stdin.encoding in interactive
modes like non-interactive modes.  This allows for non-latin-1 users
to write unicode strings directly and sets Japanese users free from
weird manual escaping <wink> in shift_jis environments.
(Reviewed by Martin v. Loewis)

20 years agoAdd a missing decref -- PyErr_SetObject increfs the 'object'!
Michael W. Hudson [Wed, 4 Aug 2004 14:59:00 +0000 (14:59 +0000)]
Add a missing decref -- PyErr_SetObject increfs the 'object'!

20 years agoFix a leak of a reference on None.
Michael W. Hudson [Wed, 4 Aug 2004 14:33:28 +0000 (14:33 +0000)]
Fix a leak of a reference on None.

20 years agoTo ever run this test "you must import TestSkipped" from the right
Michael W. Hudson [Wed, 4 Aug 2004 14:22:56 +0000 (14:22 +0000)]
To ever run this test "you must import TestSkipped" from the right
place! (can that please be it for silly mistakes in this file? :-) I
know I started it, but...).

Also, rearrangements to run repeatedly.

20 years agoBump buildno for 2.4a2
Martin v. Löwis [Wed, 4 Aug 2004 13:42:43 +0000 (13:42 +0000)]
Bump buildno for 2.4a2

20 years agoRevert 2.312; turns out interning the file name did do some good (reducing
Michael W. Hudson [Wed, 4 Aug 2004 10:26:08 +0000 (10:26 +0000)]
Revert 2.312; turns out interning the file name did do some good (reducing
.pyc size) after all.

20 years agoUpdated Logger.log() docstring to clarify that lvl parameter should be an integer.
Vinay Sajip [Wed, 4 Aug 2004 08:38:08 +0000 (08:38 +0000)]
Updated Logger.log() docstring to clarify that lvl parameter should be an integer.

20 years agoUpdated Logger.log() documentation to clarify that lvl parameter should be an integer.
Vinay Sajip [Wed, 4 Aug 2004 08:36:44 +0000 (08:36 +0000)]
Updated Logger.log() documentation to clarify that lvl parameter should be an integer.