]> git.ipfire.org Git - thirdparty/Python/cpython.git/log
thirdparty/Python/cpython.git
17 years agoPrepare for 2.3.7c1. v2.3.7c1
Martin v. Löwis [Sun, 2 Mar 2008 20:39:32 +0000 (20:39 +0000)] 
Prepare for 2.3.7c1.

17 years agoBackport of r61180:
Martin v. Löwis [Sun, 2 Mar 2008 20:32:57 +0000 (20:32 +0000)] 
Backport of r61180:
Added checks for integer overflows, contributed by Google. Some are
only available if asserts are left in the code, in cases where they
can't be triggered from Python code.

17 years agoRevert r52798, r52803, r52824, r54342, as they don't fix
Martin v. Löwis [Sun, 2 Mar 2008 17:15:58 +0000 (17:15 +0000)] 
Revert r52798, r52803, r52824, r54342, as they don't fix
security issues.

18 years agoSF patch #1556895; Typo in encoding name in email package.
Barry Warsaw [Tue, 13 Mar 2007 18:16:52 +0000 (18:16 +0000)] 
SF patch #1556895; Typo in encoding name in email package.

Patch supplied by Guillaume Rousse.

18 years ago[Jython patch #1599050] Avoid crash when os module lacks fstat()
Andrew M. Kuchling [Wed, 22 Nov 2006 19:50:21 +0000 (19:50 +0000)] 
[Jython patch #1599050] Avoid crash when os module lacks fstat()

18 years agoJython compatibility fix: if uu.decode() opened its output file, be sure to
Andrew M. Kuchling [Mon, 20 Nov 2006 13:40:36 +0000 (13:40 +0000)] 
Jython compatibility fix: if uu.decode() opened its output file, be sure to
close it.

18 years agoAdd extra SHA tests
Andrew M. Kuchling [Mon, 20 Nov 2006 13:31:37 +0000 (13:31 +0000)] 
Add extra SHA tests

19 years agoTagging for release of Python 2.3.6
Anthony Baxter [Wed, 1 Nov 2006 07:09:04 +0000 (07:09 +0000)] 
Tagging for release of Python 2.3.6

19 years agoupdate patchlevel.h correctly, this time v2.3.6
Anthony Baxter [Wed, 1 Nov 2006 07:07:27 +0000 (07:07 +0000)] 
update patchlevel.h correctly, this time

19 years agoupdates for 2.3.6 final
Anthony Baxter [Wed, 1 Nov 2006 07:04:56 +0000 (07:04 +0000)] 
updates for 2.3.6 final

19 years agoTagging for release of Python 2.3.6c1
Anthony Baxter [Mon, 23 Oct 2006 15:46:01 +0000 (15:46 +0000)] 
Tagging for release of Python 2.3.6c1

19 years agoset svn:ignore. anal retentiveness, makes it easier to spot actual problems v2.3.6c1
Anthony Baxter [Mon, 23 Oct 2006 15:38:59 +0000 (15:38 +0000)] 
set svn:ignore. anal retentiveness, makes it easier to spot actual problems

19 years agopreparing for 2.3.6c1
Anthony Baxter [Mon, 23 Oct 2006 15:27:22 +0000 (15:27 +0000)] 
preparing for 2.3.6c1

19 years agopatch for PSF-2006-001.
Anthony Baxter [Mon, 23 Oct 2006 15:23:22 +0000 (15:23 +0000)] 
patch for PSF-2006-001.

19 years agoFix the tests to work with Python 2.1, which email 2.5 must do.
Barry Warsaw [Wed, 26 Jul 2006 03:55:09 +0000 (03:55 +0000)] 
Fix the tests to work with Python 2.1, which email 2.5 must do.

19 years agoBack port r50693 and r50754 from the trunk (and 2.4 branch):
Barry Warsaw [Tue, 25 Jul 2006 13:06:56 +0000 (13:06 +0000)] 
Back port r50693 and r50754 from the trunk (and 2.4 branch):

decode_rfc2231(): Be more robust against buggy RFC 2231 encodings.
Specifically, instead of raising a ValueError when there is a single
tick in the parameter, simply return that the entire string unquoted, with
None for both the charset and the language.  Also, if there are more than 2
ticks in the parameter, interpret the first three parts as the standard RFC
2231 parts, then the rest of the parts as the encoded string.

More RFC 2231 improvements for the email 4.0 package.  As Mark Sapiro
rightly points out there are really two types of continued headers
defined in this RFC (i.e. "encoded" parameters with the form
"name*0*=" and unencoded parameters with the form "name*0="), but we
were were handling them both the same way and that isn't correct.

This patch should be much more RFC compliant in that only encoded
params are %-decoded and the charset/language information is only
extract if there are any encoded params in the segments.  If there are
no encoded params then the RFC says that there will be no
charset/language parts.

Note however that this will change the return value for
Message.get_param() in some cases.  For example, whereas before if you
had all unencoded param continuations you would have still gotten a
3-tuple back from this method (with charset and language == None), you
will now get just a string. I don't believe this is a backward
incompatible change though because the documentation for this method
already indicates that either return value is possible and that you
must do an isinstance(val, tuple) check to discriminate between the
two.  (Yeah that API kind of sucks but we can't change /that/ without
breaking code.)

Test cases, some documentation updates, and a NEWS item accompany this
patch.

Original fewer-than-3-parts fix by Tokio Kikuchi.

Resolves SF bug # 1218081.

Also, bump the package version number to 2.5.8 for release.

19 years agoBack port from 2.4 branch:
Barry Warsaw [Mon, 1 May 2006 03:21:25 +0000 (03:21 +0000)] 
Back port from 2.4 branch:

Patch #1464708 from William McVey: fixed handling of nested comments in mail
addresses.  E.g.

"Foo ((Foo Bar)) <foo@example.com>"

Fixes for both rfc822.py and email package.

19 years agoResolve SF bug 1409403: email.Message should supress warning from uu.decode.
Barry Warsaw [Thu, 9 Feb 2006 04:03:22 +0000 (04:03 +0000)] 
Resolve SF bug 1409403: email.Message should supress warning from uu.decode.

However, the patch in that tracker item is elaborated such that the newly
included unit test pass on Python 2.1 through 2.5.  Note that Python 2.1's
uu.decode() does not have a 'quiet' argument, so we have to be sneaky.

Will port to email 3.0 (although without the backward compatible sneakiness).

19 years agoPatches to address SF bugs 1409538 (Japanese codecs in CODEC_MAP) and 1409455
Barry Warsaw [Wed, 8 Feb 2006 13:33:20 +0000 (13:33 +0000)] 
Patches to address SF bugs 1409538 (Japanese codecs in CODEC_MAP) and 1409455
(.set_payload() gives bad .get_payload() results).  Specific changes include:

Simplfy the default CODEC_MAP in Charset.py to not include the Japanese and
Korean codecs.  The names of the codecs are different depending on whether
you're using Python 2.4 and 2.5, which include the codecs by default, or
earlier Python's which provide the codecs under different names as a third
party library.  Now, we attempt to discover which (if either) is available and
populate the CODEC_MAP as appropriate.

Message.set_charset(): When the message does not already have a
Content-Transfer-Encoding header, instead of just adding the header, we also
encode the body as defined by the assigned Charset.  As before, if the
body_encoding is callable, we just call that.  If not, then we add a call to
body_encode() before setting the header.  This way, we guarantee that a
message's text payload is always encoded properly.

Remove the payload encoding code from Generator._handle_text().  With the
above patch, this would cause the body to be doubly encoded.  Doing this in
the Message class is better than only doing it in the Generator.

Added some new tests to ensure everything works correctly.  Also changed the
way the test_email_codecs.py tests get added (using the same lookup code that
the CODEC_MAP adjustments use).

This resolves both issues for email 2.5/Python 2.3.  I will patch forward to
email 3.0 for both Python 2.4 and 2.5.

19 years agoResolves SF bug #1423972.
Barry Warsaw [Sat, 4 Feb 2006 23:48:22 +0000 (23:48 +0000)] 
Resolves SF bug #1423972.

19 years agoparsedate_tz(): Return a 1 in the tm_yday field so that the value is
Barry Warsaw [Fri, 3 Feb 2006 04:41:24 +0000 (04:41 +0000)] 
parsedate_tz(): Return a 1 in the tm_yday field so that the value is
acceptable to Python 2.4's time.strftime().  This fix mirrors the behavior in
email 3.0.  That field is documented as being "not useable" so it might as
well not be buggy too <wink>.

Add a test for this behavior and update a few tests that were expecting a 0 in
this field.  After committing I will run the entire Python 2.3 test suite to
ensure this doesn't break any Python tests.

19 years agoUpdated docs for email.Message.get_filename()
Barry Warsaw [Tue, 17 Jan 2006 05:17:29 +0000 (05:17 +0000)] 
Updated docs for email.Message.get_filename()

19 years agoSF bug #1403349 solution for email 2.5; some MUAs use the 'file' parameter
Barry Warsaw [Tue, 17 Jan 2006 04:34:54 +0000 (04:34 +0000)] 
SF bug #1403349 solution for email 2.5; some MUAs use the 'file' parameter
name in the Content-Distribution header, so Message.get_filename() should fall
back to using that.  Will port both to email 3.0 and Python 2.5 trunk.

Also, bump the email package version to 2.5.7 for eventual release.  Of
course, add a test case too.

XXX Need to update the documentation.

20 years agoBegin 2.3.6 news section; record PCRE fix
Andrew M. Kuchling [Wed, 31 Aug 2005 12:58:09 +0000 (12:58 +0000)] 
Begin 2.3.6 news section; record PCRE fix

20 years agoFor reference, add tests for PCRE fix; the tests aren't run by default because I...
Andrew M. Kuchling [Wed, 31 Aug 2005 12:56:50 +0000 (12:56 +0000)] 
For reference, add tests for PCRE fix; the tests aren't run by default because I wanted to minimize upheaval to the 2.3 test suite

20 years agoApply fix for potential heap overflow in PCRE code (CAN-2005-2491)
Andrew M. Kuchling [Wed, 31 Aug 2005 12:55:21 +0000 (12:55 +0000)] 
Apply fix for potential heap overflow in PCRE code (CAN-2005-2491)

20 years ago[Bug #1194249] Fix duplicate assignment in example code
Andrew M. Kuchling [Wed, 1 Jun 2005 15:28:01 +0000 (15:28 +0000)] 
[Bug #1194249] Fix duplicate assignment in example code

20 years agoget_filename(), get_content_charset(): It's possible that the charset named in
Barry Warsaw [Fri, 29 Apr 2005 12:12:02 +0000 (12:12 +0000)] 
get_filename(), get_content_charset(): It's possible that the charset named in
an RFC 2231-style header could be bogus or unknown to Python.  In that case,
we return the the text part of the parameter undecoded.  However, in
get_content_charset(), if that is not ascii, then it is an illegal charset and
so we return failobj.

Test cases and a version bump are included.

Committing this to the Python 2.3 branch because I need to generate an email
2.5.6 release that contains these patches.  I will port these fixes to Python
2.4 and 2.5 for email 3.x.

20 years agoStarted on MacPython additions for Tiger.
Jack Jansen [Fri, 22 Apr 2005 22:38:13 +0000 (22:38 +0000)] 
Started on MacPython additions for Tiger.

20 years agocorrect ontosys url - closes 1167329
Skip Montanaro [Mon, 21 Mar 2005 19:42:52 +0000 (19:42 +0000)] 
correct ontosys url - closes 1167329

20 years agoThis commit was manufactured by cvs2svn to create tag 'r235'. v2.3.5
cvs2svn [Tue, 8 Feb 2005 13:27:52 +0000 (13:27 +0000)] 
This commit was manufactured by cvs2svn to create tag 'r235'.

20 years ago2.3.5 final
Anthony Baxter [Tue, 8 Feb 2005 13:27:52 +0000 (13:27 +0000)] 
2.3.5 final

20 years agodamn
Anthony Baxter [Tue, 8 Feb 2005 13:24:28 +0000 (13:24 +0000)] 
damn

20 years ago2.3.5 final
Anthony Baxter [Tue, 8 Feb 2005 12:57:55 +0000 (12:57 +0000)] 
2.3.5 final

20 years ago2.3.5 final
Anthony Baxter [Tue, 8 Feb 2005 11:37:01 +0000 (11:37 +0000)] 
2.3.5 final

20 years agoset release date in the docs
Fred Drake [Tue, 8 Feb 2005 06:13:18 +0000 (06:13 +0000)] 
set release date in the docs

20 years agoadd 2.3.5 to the history and licensing info
Fred Drake [Tue, 8 Feb 2005 06:01:50 +0000 (06:01 +0000)] 
add 2.3.5 to the history and licensing info

20 years agobump version number
Fred Drake [Tue, 8 Feb 2005 05:58:05 +0000 (05:58 +0000)] 
bump version number

20 years agonews update for 2.3.5 final
Anthony Baxter [Tue, 8 Feb 2005 05:57:20 +0000 (05:57 +0000)] 
news update for 2.3.5 final

20 years agosligtly strengthen unit tests for copy.py
Alex Martelli [Mon, 7 Feb 2005 12:18:26 +0000 (12:18 +0000)] 
sligtly strengthen unit tests for copy.py

20 years agotiny backport from 2.4, fix a leak in _testcapi
Alex Martelli [Mon, 7 Feb 2005 12:04:22 +0000 (12:04 +0000)] 
tiny backport from 2.4, fix a leak in _testcapi

20 years agofix bug 1114776
Alex Martelli [Sun, 6 Feb 2005 07:56:18 +0000 (07:56 +0000)] 
fix bug 1114776

20 years agofix XMLFilterBase.resolveEntity() so the caller gets the result
Fred Drake [Thu, 3 Feb 2005 17:29:32 +0000 (17:29 +0000)] 
fix XMLFilterBase.resolveEntity() so the caller gets the result
(PyXML bug #1112052)

20 years agoSecurity fix PSF-2005-001 for SimpleXMLRPCServer.py.
Guido van Rossum [Thu, 3 Feb 2005 14:59:43 +0000 (14:59 +0000)] 
Security fix PSF-2005-001 for SimpleXMLRPCServer.py.

20 years agoFix typo in verbose output for RLock when the initial lock acquisition occurs.
Brett Cannon [Thu, 27 Jan 2005 22:52:09 +0000 (22:52 +0000)] 
Fix typo in verbose output for RLock when the initial lock acquisition occurs.

Closes bug #1110998.

20 years agoPartially revert #1074011; don't try to fflush stdin.
Martin v. Löwis [Thu, 27 Jan 2005 18:55:06 +0000 (18:55 +0000)] 
Partially revert #1074011; don't try to fflush stdin.

20 years agoThis commit was manufactured by cvs2svn to create tag 'r235c1'. v2.3.5c1
cvs2svn [Wed, 26 Jan 2005 04:55:57 +0000 (04:55 +0000)] 
This commit was manufactured by cvs2svn to create tag 'r235c1'.

20 years agoadded 2.3.5
Anthony Baxter [Wed, 26 Jan 2005 04:55:57 +0000 (04:55 +0000)] 
added 2.3.5

20 years agoRecompiled after source changes.
Thomas Heller [Tue, 25 Jan 2005 22:09:01 +0000 (22:09 +0000)] 
Recompiled after source changes.

20 years agoInstaller changes for 2.3.5c1.
Thomas Heller [Tue, 25 Jan 2005 22:07:51 +0000 (22:07 +0000)] 
Installer changes for 2.3.5c1.

20 years agoBuild number for 2.3.5c1 is 61.
Thomas Heller [Tue, 25 Jan 2005 22:05:22 +0000 (22:05 +0000)] 
Build number for 2.3.5c1 is 61.

20 years agoRunning a bdist_wininst installer, built with Python 2.3, installing
Thomas Heller [Tue, 25 Jan 2005 20:46:48 +0000 (20:46 +0000)] 
Running a bdist_wininst installer, built with Python 2.3, installing
for Python 2.4 caused a segfault when post_install_script was used.

The reason was that the file handle passed to PyRun_SimpleFile() was
created with MSVCRT.DLL, but Python 2.4 uses MSVCR71.DLL.

So, I replaced PyRun_SimpleFile() with PyRun_SimpleString().  The
segfault is gone, but the output of the postinstall script doesn't
show up, because still freopen() from MSVCRT is used.

(I would be very gratefull if someone proofreads the patch, at least).

20 years agoSet PYTHONHOME before loading Python, so that sys.path will be set correctly.
Thomas Heller [Tue, 25 Jan 2005 19:47:48 +0000 (19:47 +0000)] 
Set PYTHONHOME before loading Python, so that sys.path will be set correctly.

20 years agoset the release date
Fred Drake [Tue, 25 Jan 2005 15:16:49 +0000 (15:16 +0000)] 
set the release date

20 years ago1.0.5, to match the python patchlevel
Anthony Baxter [Tue, 25 Jan 2005 13:08:15 +0000 (13:08 +0000)] 
1.0.5, to match the python patchlevel

20 years agocopy.py fixed to first lookup __copy__ from the instance being copied,
Anthony Baxter [Tue, 25 Jan 2005 12:52:18 +0000 (12:52 +0000)] 
copy.py fixed to first lookup __copy__ from the instance being copied,
rather than only looking at the type - this was broken in 2.3.

20 years agopre-release bits
Anthony Baxter [Tue, 25 Jan 2005 12:51:43 +0000 (12:51 +0000)] 
pre-release bits

20 years agohappy new year!
Anthony Baxter [Tue, 25 Jan 2005 12:39:22 +0000 (12:39 +0000)] 
happy new year!

20 years agobump version number to prepare for a release
Fred Drake [Tue, 25 Jan 2005 07:53:30 +0000 (07:53 +0000)] 
bump version number to prepare for a release

20 years agoFlush std{in,out,err} before closing it. Fixes #1074011.
Martin v. Löwis [Sun, 23 Jan 2005 09:50:32 +0000 (09:50 +0000)] 
Flush std{in,out,err} before closing it. Fixes #1074011.

20 years agoAt least one bug has been fixed in distutils, so increment the last part of
Thomas Heller [Thu, 20 Jan 2005 19:28:44 +0000 (19:28 +0000)] 
At least one bug has been fixed in distutils, so increment the last part of
the version number.  For the distutils version numbering scheme, see
http://mail.python.org/pipermail/distutils-sig/2005-January/004368.html

20 years agodistutils/command/install.py change.
Thomas Heller [Thu, 20 Jan 2005 19:22:12 +0000 (19:22 +0000)] 
distutils/command/install.py change.

20 years agoFix [ 1103844 ] fix distutils.install.dump_dirs() with negated options.
Thomas Heller [Thu, 20 Jan 2005 19:16:27 +0000 (19:16 +0000)] 
Fix [ 1103844 ] fix distutils.install.dump_dirs() with negated options.

Backport from trunk.

20 years agofix typeset of vertical bar for PDF format
Fred Drake [Wed, 19 Jan 2005 07:22:13 +0000 (07:22 +0000)] 
fix typeset of vertical bar for PDF format
(closes SF bug #1008998)

20 years agoOn OpenBSD, terminating IDLE with ctrl-c from the command line caused a
Kurt B. Kaiser [Wed, 19 Jan 2005 06:23:51 +0000 (06:23 +0000)] 
On OpenBSD, terminating IDLE with ctrl-c from the command line caused a
stuck subprocess MainThread because only the SocketThread was exiting.

M NEWS.txt
M run.py

20 years agodocumentation for the zipimport module using contributed patch
Fred Drake [Wed, 19 Jan 2005 06:03:29 +0000 (06:03 +0000)] 
documentation for the zipimport module using contributed patch
(closes SF bug #853800; markup adjusted)

20 years agoBackport configDialog.py rev 1.60:
Kurt B. Kaiser [Wed, 19 Jan 2005 05:53:32 +0000 (05:53 +0000)] 
Backport configDialog.py rev 1.60:

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.

M NEWS.txt
M configDialog.py

20 years agoThis commit was manufactured by cvs2svn to create branch
cvs2svn [Wed, 19 Jan 2005 05:42:50 +0000 (05:42 +0000)] 
This commit was manufactured by cvs2svn to create branch
'release23-maint'.

20 years agoDescription of args to IMAP4.store() in imaplib
Fred Drake [Wed, 19 Jan 2005 04:49:26 +0000 (04:49 +0000)] 
Description of args to IMAP4.store() in imaplib
(closes SF patch #1084092; modified per comments in SF)

20 years agoBackport EditorWindow rev 1.60 configHelpSourceEdit rev 1.7 15Jul04
Kurt B. Kaiser [Wed, 19 Jan 2005 04:12:38 +0000 (04:12 +0000)] 
Backport EditorWindow rev 1.60  configHelpSourceEdit rev 1.7  15Jul04

checking sys.platform for substring 'win' was breaking IDLE docs on Mac
(darwin).  Also, Mac Safari browser requires full file:// URIs.  SF 900580.

M EditorWindow.py
M NEWS.txt
M configHelpSourceEdit.py

20 years agofix a bunch of spelling errors
Fred Drake [Wed, 19 Jan 2005 03:45:39 +0000 (03:45 +0000)] 
fix a bunch of spelling errors
(closes SF patch #1104868)

20 years agoremove spurious line from widget variables example
Fred Drake [Wed, 19 Jan 2005 03:27:18 +0000 (03:27 +0000)] 
remove spurious line from widget variables example
(closes SF patch #1094815)

20 years agoBackport rpc.py rev 1.28 dating from 21Jan04
Kurt B. Kaiser [Wed, 19 Jan 2005 01:44:06 +0000 (01:44 +0000)] 
Backport rpc.py rev 1.28 dating from 21Jan04

rpc.py:SocketIO - Large modules were generating large pickles when downloaded
to the execution server.  The return of the OK response from the subprocess
initialization was interfering and causing the sending socket to be not
ready.  Add an IO ready test to fix this.  Moved the polling IO ready test
into pollpacket().

Fix typo in rpc.py, s/b "pickle.PicklingError" not "pickle.UnpicklingError".

idlever.py should be 1.0.4 to align with NEWS.txt.  There was no IDLE release
at 2.3.1 which accounts for the unsync.

M NEWS.txt
M idlever.py
M rpc.py

20 years agoBackport Patch #850977: Detect Tk on FreeBSD and OpenBSD.
Kurt B. Kaiser [Mon, 17 Jan 2005 21:07:52 +0000 (21:07 +0000)] 
Backport Patch #850977: Detect Tk on FreeBSD and OpenBSD.

20 years agoIf an extension can't be loaded, print warning and skip it instead of
Kurt B. Kaiser [Mon, 17 Jan 2005 20:34:49 +0000 (20:34 +0000)] 
If an extension can't be loaded, print warning and skip it instead of
erroring out.

20 years agoImprove error handling when .idlerc can't be created. This is a partial
Kurt B. Kaiser [Mon, 17 Jan 2005 20:06:47 +0000 (20:06 +0000)] 
Improve error handling when .idlerc can't be created.  This is a partial
backport of configHandler.py, Revision 1.36,  11Jan05.

20 years agoupdate in readiness for 2.3.5 release
Andrew MacIntyre [Mon, 17 Jan 2005 13:06:27 +0000 (13:06 +0000)] 
update in readiness for 2.3.5 release

20 years agobackport: make thread stack size compile-time tunable on OS/2, incr main stack
Andrew MacIntyre [Mon, 17 Jan 2005 12:22:38 +0000 (12:22 +0000)] 
backport: make thread stack size compile-time tunable on OS/2, incr main stack

20 years agobackport: make thread stack size compile-time tunable on OS/2
Andrew MacIntyre [Mon, 17 Jan 2005 12:21:28 +0000 (12:21 +0000)] 
backport: make thread stack size compile-time tunable on OS/2

20 years agoBackport of fixes for #887242 and #1097739:
Jack Jansen [Tue, 11 Jan 2005 13:49:02 +0000 (13:49 +0000)] 
Backport of fixes for #887242 and #1097739:

If $MACOSX_DEPLOYMENT_TARGET is set, and >= 10.3, during configure we
setup extensions to link with dynamic lookup. We also record the
value in the Makefile, and distutils uses the same value to build extension
modules. If MACOSX_DEPLOYMENT_TARGET is not set it defaults to the
current OSX version.

If we cannot use -undefined dynamic_lookup (such as on 10.2 or earlier)
we link extensions directly against the dynamic library in the framework in
stead of against the framework. This will fix building extensions for 2.3
after 2.4 has been installed too.

20 years agoBackport latest version of whatsnew23: fix typos, remove a GC-related item; add an...
Andrew M. Kuchling [Fri, 7 Jan 2005 14:43:39 +0000 (14:43 +0000)] 
Backport latest version of whatsnew23: fix typos, remove a GC-related item; add an item

20 years agoBackport of 1.38:
Jack Jansen [Fri, 7 Jan 2005 13:48:53 +0000 (13:48 +0000)] 
Backport of 1.38:
Allow relative URLs for included databases and packages.

20 years agoBackport of 1.20:
Jack Jansen [Fri, 7 Jan 2005 12:57:08 +0000 (12:57 +0000)] 
Backport of 1.20:
Workaround for the fact that urllib2 doesn't default to "file:" urls.

20 years agoMerging the current development changes for the idle wrapper.
Sean Reifschneider [Wed, 5 Jan 2005 04:59:29 +0000 (04:59 +0000)] 
Merging the current development changes for the idle wrapper.

20 years agoMac news items.
Jack Jansen [Tue, 4 Jan 2005 16:07:04 +0000 (16:07 +0000)] 
Mac news items.

20 years agoBackport of 1.55 and 1.57:
Jack Jansen [Tue, 4 Jan 2005 16:05:33 +0000 (16:05 +0000)] 
Backport of 1.55 and 1.57:
- Added quotes around the destroot arguments, so empty destroot works.
- Run fixapplepython23 script to check whether Apple's Python 2.3 needs
  to be patched.

20 years agoFix for #1076490 submitted by Neil Mayhew: a serious typo means
Jack Jansen [Tue, 4 Jan 2005 15:25:48 +0000 (15:25 +0000)] 
Fix for #1076490 submitted by Neil Mayhew: a serious typo means
BuildApplication has never worked in 2.3. I guess that shows how
popular MacOS9 is:-)

20 years agoAdded one-liner about pimp 0.5
Jack Jansen [Mon, 3 Jan 2005 16:00:28 +0000 (16:00 +0000)] 
Added one-liner about pimp 0.5

20 years agoBackport of 1.37:
Jack Jansen [Mon, 3 Jan 2005 15:46:30 +0000 (15:46 +0000)] 
Backport of 1.37:
- Added an "installer" flavor, which uses the "open" command to install
  something (overridable through Install-command entry)
- Hidden status is now determined by flavor == hidden, not by
  missing Download-URL. Hidden packages behave like installer packages.
- Made some error messages a bit more understandable.

Because there's new functionality the version has been upped to 0.5.

20 years agoGetting things in place for the MacPython additions build 3.
Jack Jansen [Sun, 2 Jan 2005 23:19:40 +0000 (23:19 +0000)] 
Getting things in place for the MacPython additions build 3.

20 years agoBackported from the trunk.
Jack Jansen [Sun, 2 Jan 2005 23:18:07 +0000 (23:18 +0000)] 
Backported from the trunk.

20 years agoThis commit was manufactured by cvs2svn to create branch
cvs2svn [Sun, 2 Jan 2005 23:18:07 +0000 (23:18 +0000)] 
This commit was manufactured by cvs2svn to create branch
'release23-maint'.

20 years agoAdded a band-aid to make this script work with the older 2.3.0
Jack Jansen [Sun, 2 Jan 2005 23:17:06 +0000 (23:17 +0000)] 
Added a band-aid to make this script work with the older 2.3.0
buildtools if no destroot option is given. That makes life for the
additions installer a lot simpler.

20 years agoBackport:
Jack Jansen [Fri, 31 Dec 2004 10:44:41 +0000 (10:44 +0000)] 
Backport:
Fix for #1091468: DESTROOTed frameworkinstalls fail. Added a --destroot
option to various tools, and do the right thing when we're doing a destroot
install.

20 years agoSpaces in stead of tabs.
Jack Jansen [Thu, 30 Dec 2004 22:43:29 +0000 (22:43 +0000)] 
Spaces in stead of tabs.

20 years agoPatch 728330: Fix compilation problems on IRIX.
Martin v. Löwis [Thu, 30 Dec 2004 13:52:17 +0000 (13:52 +0000)] 
Patch 728330: Fix compilation problems on IRIX.

20 years agoBackport of 1.36:
Jack Jansen [Tue, 28 Dec 2004 21:53:15 +0000 (21:53 +0000)] 
Backport of 1.36:
- getDefaultDatabase wasn't listed in __all__.
- using a different database for non-final releases should only be done
  for X.Y.0. Non-final micro releases can use the default database just fine,
  as they are required to be backward compatible.

20 years ago[Bug #1083110] calling .flush() on decompress objects causes a segfault due to an...
Andrew M. Kuchling [Tue, 28 Dec 2004 20:31:27 +0000 (20:31 +0000)] 
[Bug #1083110] calling .flush() on decompress objects causes a segfault due to an uninitialized pointer: fixes the problem and adds a test case

20 years agoUse zlib test suite from the trunk (rev. 2.67) on the 2.3 maintenance branch
Andrew M. Kuchling [Tue, 28 Dec 2004 20:26:00 +0000 (20:26 +0000)] 
Use zlib test suite from the trunk (rev. 2.67) on the 2.3 maintenance branch

20 years ago- Changed version numbers to get ready for 2.3.5
Jack Jansen [Mon, 27 Dec 2004 16:41:57 +0000 (16:41 +0000)] 
- Changed version numbers to get ready for 2.3.5
- Various of the mods to build the 3rd MacPython additions (but not all).