]> granicus.if.org Git - python/log
python
12 years agoFix sporadic failure of test_time.test_process_time() on Windows
Victor Stinner [Fri, 1 Jun 2012 20:45:23 +0000 (22:45 +0200)]
Fix sporadic failure of test_time.test_process_time() on Windows

Use a threshold of 20 ms instead of 10 ms.

12 years ago#14957: clarify splitlines docs.
R David Murray [Fri, 1 Jun 2012 20:20:26 +0000 (16:20 -0400)]
#14957: clarify splitlines docs.

Initial patch by Michael Driscoll, I added the example.

12 years ago#14957: clarify splitlines docs.
R David Murray [Fri, 1 Jun 2012 20:19:36 +0000 (16:19 -0400)]
#14957: clarify splitlines docs.

Initial patch by Michael Driscoll, I added the example.

12 years agomerge heads
Sandro Tosi [Fri, 1 Jun 2012 18:25:36 +0000 (20:25 +0200)]
merge heads

12 years agoIssue #14968: merge with 3.2
Sandro Tosi [Fri, 1 Jun 2012 18:23:46 +0000 (20:23 +0200)]
Issue #14968: merge with 3.2

12 years agoIssue #14968: set 'Inplace Operators' as subsection; patch by Lars Buitinck
Sandro Tosi [Fri, 1 Jun 2012 18:23:20 +0000 (20:23 +0200)]
Issue #14968: set 'Inplace Operators' as subsection; patch by Lars Buitinck

12 years agocheck return for error
Benjamin Peterson [Fri, 1 Jun 2012 18:18:22 +0000 (11:18 -0700)]
check return for error

12 years ago#14814: Fix errror message creation in ipaddress.collapse_addresses
Hynek Schlawack [Fri, 1 Jun 2012 18:12:17 +0000 (20:12 +0200)]
#14814: Fix errror message creation in ipaddress.collapse_addresses

12 years agoClose #14969: Improve the handling of exception chaining in contextlib.ExitStack
Nick Coghlan [Fri, 1 Jun 2012 12:48:32 +0000 (22:48 +1000)]
Close #14969: Improve the handling of exception chaining in contextlib.ExitStack

12 years ago#14814: Remove 2.x's new-style classes syntax from ipaddress
Hynek Schlawack [Fri, 1 Jun 2012 09:48:32 +0000 (11:48 +0200)]
#14814: Remove 2.x's new-style classes syntax from ipaddress

12 years agoword.digits are always initialized before use in the Taylor series loop,
Stefan Krah [Fri, 1 Jun 2012 08:58:16 +0000 (10:58 +0200)]
word.digits are always initialized before use in the Taylor series loop,
but this is more readable.

12 years agoIssue #14007: implement doctype() method calling in XMLParser of _elementtree.
Eli Bendersky [Fri, 1 Jun 2012 08:32:34 +0000 (11:32 +0300)]
Issue #14007: implement doctype() method calling in XMLParser of _elementtree.
Includes exposing a doctype handler from expat through pyexpat.

12 years agoWe're always building _elementtree with USE_PYEXPAT_CAPI, so the #ifdefs in
Eli Bendersky [Fri, 1 Jun 2012 06:48:37 +0000 (09:48 +0300)]
We're always building _elementtree with USE_PYEXPAT_CAPI, so the #ifdefs in
the code are unnecessary.

12 years agoReformat two issue numbers
Brian Curtin [Fri, 1 Jun 2012 05:07:28 +0000 (00:07 -0500)]
Reformat two issue numbers

12 years agofix trailing whitespace
Eli Bendersky [Fri, 1 Jun 2012 04:15:00 +0000 (07:15 +0300)]
fix trailing whitespace

12 years agoIssue #14007: make XMLParser a real subclassable type exported from _elementtree...
Eli Bendersky [Fri, 1 Jun 2012 04:13:08 +0000 (07:13 +0300)]
Issue #14007: make XMLParser a real subclassable type exported from _elementtree. +cleanups

12 years ago#14814: Remove stale __hex__ method from ipaddress
Hynek Schlawack [Thu, 31 May 2012 22:20:13 +0000 (00:20 +0200)]
#14814: Remove stale __hex__ method from ipaddress

Obsolete 2.x method.

12 years agoDon't use metaclasses when class decorators can do the job.
R David Murray [Thu, 31 May 2012 22:00:45 +0000 (18:00 -0400)]
Don't use metaclasses when class decorators can do the job.

Thanks to Nick Coghlan for pointing out that I'd forgotten about class
decorators.

12 years agoPost-release update.
Georg Brandl [Thu, 31 May 2012 20:41:51 +0000 (22:41 +0200)]
Post-release update.

12 years agoVS 2010 has the AMD64 redist files in VC\redist\x64.
Martin v. Löwis [Thu, 31 May 2012 19:58:21 +0000 (21:58 +0200)]
VS 2010 has the AMD64 redist files in VC\redist\x64.

12 years agoUpdate to Tcl/Tk 8.5.11.
Martin v. Löwis [Thu, 31 May 2012 19:53:36 +0000 (21:53 +0200)]
Update to Tcl/Tk 8.5.11.

12 years agoUse workctx instead of ctx for cosmetic reasons. Also zero-pad the result
Stefan Krah [Thu, 31 May 2012 18:49:24 +0000 (20:49 +0200)]
Use workctx instead of ctx for cosmetic reasons. Also zero-pad the result
in the simple path (not correctly rounded but faster).

12 years agoImprove Underflow handling in the correct-rounding loop. The case for
Stefan Krah [Thu, 31 May 2012 18:01:05 +0000 (20:01 +0200)]
Improve Underflow handling in the correct-rounding loop. The case for
Underflow to zero hasn't changed: _mpd_qexp() internally uses MIN_EMIN,
so the result would also underflow to zero for all emin > MIN_EMIN.

In case digits are left, the informal argument is as follows: Underflow can
occur only once in the last multiplication of the power stage (in the Horner
stage Underflow provably cannot occur, and if Underflow occurred twice in
the power stage, the result would underflow to zero on the second occasion).

Since there is no double rounding during Underflow, the effective work
precision is now 1 <= result->digits < prec. It can be shown by a somewhat
tedious argument that abs(result - e**x) < ulp(result, result->digits).

Therefore the correct rounding loop now uses ulp(result, result->digits)
to generate the bounds for e**x in case of Underflow.

12 years agoIssue #14962: merge
Ned Deily [Thu, 31 May 2012 16:58:08 +0000 (09:58 -0700)]
Issue #14962: merge

12 years agoIssue #14962: Update text coloring in IDLE shell window after changing
Ned Deily [Thu, 31 May 2012 16:17:29 +0000 (09:17 -0700)]
Issue #14962: Update text coloring in IDLE shell window after changing
options.  Patch by Roger Serwy.

12 years agoIssue #14952: Fix incorrect output dll names for win64/debug builds, causing
Kristjan Valur Jonsson [Thu, 31 May 2012 15:09:21 +0000 (15:09 +0000)]
Issue #14952: Fix incorrect output dll names for win64/debug builds, causing
the dll importer on windows to fail.

12 years agoImprove comments.
Stefan Krah [Thu, 31 May 2012 14:21:34 +0000 (16:21 +0200)]
Improve comments.

12 years agoMerge.
Stefan Krah [Thu, 31 May 2012 14:03:49 +0000 (16:03 +0200)]
Merge.

12 years agoClose #14963: Use an iterative algorithm in contextlib.ExitStack.__exit__ (Patch...
Nick Coghlan [Thu, 31 May 2012 14:00:38 +0000 (00:00 +1000)]
Close #14963: Use an iterative algorithm in contextlib.ExitStack.__exit__ (Patch by Alon Horev)

12 years agoPad the result with zeros just before the final rounding.
Stefan Krah [Thu, 31 May 2012 14:00:21 +0000 (16:00 +0200)]
Pad the result with zeros just before the final rounding.

12 years agoIssue #14963: Added test cases for contextlib.ExitStack exception handling behaviour...
Nick Coghlan [Thu, 31 May 2012 13:49:26 +0000 (23:49 +1000)]
Issue #14963: Added test cases for contextlib.ExitStack exception handling behaviour (Initial patch by Alon Horev)

12 years agoDo not clobber existing flags.
Stefan Krah [Thu, 31 May 2012 13:09:27 +0000 (15:09 +0200)]
Do not clobber existing flags.

12 years agoTypo fix
Nick Coghlan [Thu, 31 May 2012 12:34:59 +0000 (22:34 +1000)]
Typo fix

12 years agoAdditional ExitStack examples, and a few other cleanups for the ExitStack docs
Nick Coghlan [Thu, 31 May 2012 12:17:08 +0000 (22:17 +1000)]
Additional ExitStack examples, and a few other cleanups for the ExitStack docs

12 years agoNull merge from 3.2.
Vinay Sajip [Thu, 31 May 2012 11:38:55 +0000 (12:38 +0100)]
Null merge from 3.2.

12 years agoAdded test skip under Windows, as not applicable there.
Vinay Sajip [Thu, 31 May 2012 11:37:04 +0000 (12:37 +0100)]
Added test skip under Windows, as not applicable there.

12 years agoIssue #14909: A number of places were using PyMem_Realloc() apis and
Kristjan Valur Jonsson [Thu, 31 May 2012 09:37:31 +0000 (09:37 +0000)]
Issue #14909: A number of places were using PyMem_Realloc() apis and
PyObject_GC_Resize() with incorrect error handling.  In case of errors,
the original object would be leaked.  This checkin fixes those cases.

12 years agoMake parameterized tests in email less hackish.
R David Murray [Thu, 31 May 2012 01:53:40 +0000 (21:53 -0400)]
Make parameterized tests in email less hackish.

Or perhaps more hackish, depending on your perspective.  But at least this
way it is now possible to run the individual tests using the unittest CLI.

12 years agoMerge heads.
Georg Brandl [Wed, 30 May 2012 20:04:57 +0000 (22:04 +0200)]
Merge heads.

12 years agoAdded tag v3.3.0a4 for changeset 7c51388a3aa7
Georg Brandl [Wed, 30 May 2012 20:04:40 +0000 (22:04 +0200)]
Added tag v3.3.0a4 for changeset 7c51388a3aa7

12 years agoBump version to 3.3.0a4. v3.3.0a4
Georg Brandl [Wed, 30 May 2012 20:04:31 +0000 (22:04 +0200)]
Bump version to 3.3.0a4.

12 years agoUpdate pydoc topics and fix new suspicious markup.
Georg Brandl [Wed, 30 May 2012 20:03:20 +0000 (22:03 +0200)]
Update pydoc topics and fix new suspicious markup.

12 years agoIssue #14007: implemented the 'element_factory' feature of TreeBuilder in
Eli Bendersky [Wed, 30 May 2012 14:57:50 +0000 (17:57 +0300)]
Issue #14007: implemented the 'element_factory' feature of TreeBuilder in
_elementtree, with a test.

12 years agoClose #14947: add missing cross-reference to Language Definition from the new dynamic...
Nick Coghlan [Wed, 30 May 2012 12:17:30 +0000 (22:17 +1000)]
Close #14947: add missing cross-reference to Language Definition from the new dynamic type creation functions. Also cleaned up the general wording of the docs

12 years ago#14796: fix failure of new calendar test on windows.
R David Murray [Wed, 30 May 2012 12:10:54 +0000 (08:10 -0400)]
#14796: fix failure of new calendar test on windows.

12 years agoClose #14690: Use monotonic clock instead of system clock in the sched,
Victor Stinner [Wed, 30 May 2012 11:30:32 +0000 (13:30 +0200)]
Close #14690: Use monotonic clock instead of system clock in the sched,
subprocess and trace modules.

12 years agoIssue #14958: Change IDLE systax highlighting to recognize all string and
Ned Deily [Wed, 30 May 2012 05:55:43 +0000 (22:55 -0700)]
Issue #14958: Change IDLE systax highlighting to recognize all string and
byte literals supported in Python 3.3.

12 years agoMerge 3.2 news item
Brian Curtin [Tue, 29 May 2012 23:44:17 +0000 (18:44 -0500)]
Merge 3.2 news item

12 years agoAdd news item for #14943
Brian Curtin [Tue, 29 May 2012 23:41:30 +0000 (18:41 -0500)]
Add news item for #14943

12 years agoFix #14943. Merge 3.2
Brian Curtin [Tue, 29 May 2012 23:36:40 +0000 (18:36 -0500)]
Fix #14943. Merge 3.2

12 years agoFix #14943. Update the proper default value and list the proper argument names in...
Brian Curtin [Tue, 29 May 2012 23:34:45 +0000 (18:34 -0500)]
Fix #14943. Update the proper default value and list the proper argument names in the explanation.

12 years agoReapplied skip logic for test on Windows, which appears to have been lost during...
Vinay Sajip [Tue, 29 May 2012 21:48:10 +0000 (22:48 +0100)]
Reapplied skip logic for test on Windows, which appears to have been lost during a merge.

12 years agoUpdated test to reflect renamed file.
Vinay Sajip [Tue, 29 May 2012 21:29:50 +0000 (22:29 +0100)]
Updated test to reflect renamed file.

12 years agoIssue #10997: merge from 3.2
Ned Deily [Tue, 29 May 2012 17:51:38 +0000 (10:51 -0700)]
Issue #10997: merge from 3.2

12 years agoIssue #10997: Prevent a duplicate entry in IDLE's "Recent Files" menu.
Ned Deily [Tue, 29 May 2012 17:43:36 +0000 (10:43 -0700)]
Issue #10997: Prevent a duplicate entry in IDLE's "Recent Files" menu.

12 years ago#14796: improve calendar test coverage.
R David Murray [Tue, 29 May 2012 16:55:05 +0000 (12:55 -0400)]
#14796: improve calendar test coverage.

Patch by Oleg Plakhotnyuk.

12 years agoIssue #14744: Fix compilation on Windows (part 2)
Victor Stinner [Tue, 29 May 2012 16:53:56 +0000 (18:53 +0200)]
Issue #14744: Fix compilation on Windows (part 2)

12 years agoIssue #14744: Fix compilation on Windows
Victor Stinner [Tue, 29 May 2012 16:51:10 +0000 (18:51 +0200)]
Issue #14744: Fix compilation on Windows

12 years ago#10839: add new test file that was omitted from checkin
R David Murray [Tue, 29 May 2012 16:31:11 +0000 (12:31 -0400)]
#10839: add new test file that was omitted from checkin

12 years ago#10839: raise an error on add of duplicate unique headers in new email policies
R David Murray [Tue, 29 May 2012 13:14:44 +0000 (09:14 -0400)]
#10839: raise an error on add of duplicate unique headers in new email policies

This feature was supposed to be part of the initial email6 checkin, but it got
lost in my big refactoring.

In this patch I'm not providing an easy way to turn off the errors, but they
only happen when a header is added programmatically, and it is almost never
the right thing to do to allow the duplicate to be added.  An application that
needs to add duplicates of unique headers can create a policy subclass to
allow it.

12 years agomerge heads
Eli Bendersky [Tue, 29 May 2012 12:45:40 +0000 (15:45 +0300)]
merge heads

12 years agoIssue #14007: make TreeBuilder an actual type exposed from _elementtree, and subclass...
Eli Bendersky [Tue, 29 May 2012 12:45:16 +0000 (15:45 +0300)]
Issue #14007: make TreeBuilder an actual type exposed from _elementtree, and subclassable.

12 years agoMerged upstream changes.
Vinay Sajip [Tue, 29 May 2012 11:53:00 +0000 (12:53 +0100)]
Merged upstream changes.

12 years agoAdded pyvenv docs to 'Python Setup and Usage' section.
Vinay Sajip [Tue, 29 May 2012 11:52:14 +0000 (12:52 +0100)]
Added pyvenv docs to 'Python Setup and Usage' section.

12 years agoIssue #14744: Use the new _PyUnicodeWriter internal API to speed up str%args and...
Victor Stinner [Tue, 29 May 2012 10:57:52 +0000 (12:57 +0200)]
Issue #14744: Use the new _PyUnicodeWriter internal API to speed up str%args and str.format(args)

 * Formatting string, int, float and complex use the _PyUnicodeWriter API. It
   avoids a temporary buffer in most cases.
 * Add _PyUnicodeWriter_WriteStr() to restore the PyAccu optimization: just
   keep a reference to the string if the output is only composed of one string
 * Disable overallocation when formatting the last argument of str%args and
   str.format(args)
 * Overallocation allocates at least 100 characters: add min_length attribute
   to the _PyUnicodeWriter structure
 * Add new private functions: _PyUnicode_FastCopyCharacters(),
   _PyUnicode_FastFill() and _PyUnicode_FromASCII()

The speed up is around 20% in average.

12 years agoPyArg_Parse*("U"): ensure that the Unicode string is ready
Victor Stinner [Tue, 29 May 2012 10:30:29 +0000 (12:30 +0200)]
PyArg_Parse*("U"): ensure that the Unicode string is ready

12 years agoUse Python 3.x-style keyword only arg in Array()
Richard Oudkerk [Tue, 29 May 2012 11:01:47 +0000 (12:01 +0100)]
Use Python 3.x-style keyword only arg in Array()

Previously a Python 2.x compatible hack was used for
multiprocessing.sharedctypes.Array().  Also the documented
signature was wrong.

12 years agoRemove __getslice__, __setslice__, __delslice__ methods from proxies
Richard Oudkerk [Tue, 29 May 2012 11:01:45 +0000 (12:01 +0100)]
Remove __getslice__, __setslice__, __delslice__ methods from proxies

Proxy classes in multiprocessing do not need these methods in Python 3.x.

12 years agoRemoved pysetup3.exe executable.
Vinay Sajip [Tue, 29 May 2012 10:12:43 +0000 (11:12 +0100)]
Removed pysetup3.exe executable.

12 years ago#14835: Make plistlib output empty arrays & dicts like OS X
Hynek Schlawack [Tue, 29 May 2012 10:04:54 +0000 (12:04 +0200)]
#14835: Make plistlib output empty arrays & dicts like OS X

Patch by Sidney San Martín.

12 years agoIssue 14814: Add namespaces keyword arg to find(*) methods in _elementtree.
Eli Bendersky [Tue, 29 May 2012 03:02:56 +0000 (06:02 +0300)]
Issue 14814: Add namespaces keyword arg to find(*) methods in _elementtree.
Add attrib keyword to Element and SubElement in _elementtree.
Patch developed with Ezio Melotti.

12 years agoRefactor test_email/test_pickleable and add tests for date headers
R David Murray [Tue, 29 May 2012 01:09:04 +0000 (21:09 -0400)]
Refactor test_email/test_pickleable and add tests for date headers

12 years agoRegularize test_email/test_headerregistry's references to policy.
R David Murray [Tue, 29 May 2012 00:22:37 +0000 (20:22 -0400)]
Regularize test_email/test_headerregistry's references to policy.

12 years agoRefactor test_email/test_defect_handling.
R David Murray [Tue, 29 May 2012 00:14:10 +0000 (20:14 -0400)]
Refactor test_email/test_defect_handling.

12 years agoIssue #14930: Make memoryview objects weakrefable.
Richard Oudkerk [Mon, 28 May 2012 20:35:09 +0000 (21:35 +0100)]
Issue #14930: Make memoryview objects weakrefable.

12 years agoIssue #14775: Fix a potential quadratic dict build-up due to the garbage collector...
Antoine Pitrou [Mon, 28 May 2012 20:23:42 +0000 (22:23 +0200)]
Issue #14775: Fix a potential quadratic dict build-up due to the garbage collector repeatedly trying to untrack dicts.
Additional comments by Tim Silk.

12 years agoIssue #14775: Fix a potential quadratic dict build-up due to the garbage collector...
Antoine Pitrou [Mon, 28 May 2012 20:22:34 +0000 (22:22 +0200)]
Issue #14775: Fix a potential quadratic dict build-up due to the garbage collector repeatedly trying to untrack dicts.
Additional comments by Tim Silk.

12 years agoIssue #9041: raised exception is misleading
Meador Inge [Mon, 28 May 2012 19:47:53 +0000 (14:47 -0500)]
Issue #9041: raised exception is misleading

An issue in ctypes.c_longdouble, ctypes.c_double, and ctypes.c_float that
caused an incorrect exception to be returned in the case of overflow has been
fixed.

12 years agoIssue #9041: raised exception is misleading
Meador Inge [Mon, 28 May 2012 19:21:16 +0000 (14:21 -0500)]
Issue #9041: raised exception is misleading

An issue in ctypes.c_longdouble, ctypes.c_double, and ctypes.c_float that
caused an incorrect exception to be returned in the case of overflow has been
fixed.

12 years agoFixed typo.
Vinay Sajip [Mon, 28 May 2012 16:01:17 +0000 (17:01 +0100)]
Fixed typo.

12 years agoTweaked tests to use launcher executable name on OS X.
Vinay Sajip [Mon, 28 May 2012 15:34:47 +0000 (16:34 +0100)]
Tweaked tests to use launcher executable name on OS X.

12 years agoAdded python3 symlink; removed Distribute-related code, docs and comments; changed...
Vinay Sajip [Mon, 28 May 2012 15:33:01 +0000 (16:33 +0100)]
Added python3 symlink; removed Distribute-related code, docs and comments; changed Mac OS X computation to determine framework builds.

12 years agoMerge from 3.2
Nick Coghlan [Mon, 28 May 2012 12:36:07 +0000 (22:36 +1000)]
Merge from 3.2

12 years agoIssue #14443: Tell rpmbuild to use the correct version of Python
Nick Coghlan [Mon, 28 May 2012 12:34:46 +0000 (22:34 +1000)]
Issue #14443: Tell rpmbuild to use the correct version of Python

12 years agoIssue #14660: Install namespace_pkgs test directories and files.
Ned Deily [Mon, 28 May 2012 05:34:33 +0000 (22:34 -0700)]
Issue #14660: Install namespace_pkgs test directories and files.

12 years agoMerge 3.2 #14929
Terry Jan Reedy [Mon, 28 May 2012 03:06:14 +0000 (23:06 -0400)]
Merge 3.2 #14929

12 years agoIssue14929: Stop Idle 3.x from closing on Unicode decode errors when grepping.
Terry Jan Reedy [Mon, 28 May 2012 02:56:49 +0000 (22:56 -0400)]
Issue14929: Stop Idle 3.x from closing on Unicode decode errors when grepping.
Patch by Roger Serwy.

12 years ago#12515: email now registers a defect if the MIME end boundary is missing.
R David Murray [Mon, 28 May 2012 02:20:42 +0000 (22:20 -0400)]
#12515: email now registers a defect if the MIME end boundary is missing.

This commit also restores the news item for 167256 that it looks like
Terry inadvertently deleted.  (Either that, or I don't understand
now merging works...which is equally possible.)

12 years agomerge heads
Terry Jan Reedy [Mon, 28 May 2012 01:54:28 +0000 (21:54 -0400)]
merge heads

12 years agoMerge 3.2 closes #12510
Terry Jan Reedy [Mon, 28 May 2012 01:39:39 +0000 (21:39 -0400)]
Merge 3.2 closes #12510

12 years agoIssue12510: Attempting to get invalid tooltip no longer closes Idle.
Terry Jan Reedy [Mon, 28 May 2012 01:29:17 +0000 (21:29 -0400)]
Issue12510: Attempting to get invalid tooltip no longer closes Idle.
Original patch by Roger Serwy.

12 years agoNews item for #1672568.
R David Murray [Mon, 28 May 2012 01:27:31 +0000 (21:27 -0400)]
News item for #1672568.

12 years ago#1672568: email now registers defects for base64 payload format errors.
R David Murray [Mon, 28 May 2012 01:23:34 +0000 (21:23 -0400)]
#1672568: email now registers defects for base64 payload format errors.

Which also means that it is now producing *something* for any base64
payload, which is what leads to the couple of older test changes in
test_email.  This is a slightly backward incompatible behavior change,
but the new behavior is so much more useful than the old (you can now
*reliably* detect errors, and any program that was detecting errors by
sniffing for a base64 return from get_payload(decode=True) and then doing
its own error-recovery decode will just get the error-recovery decode
right away).  So this seems to me to be worth the small risk inherent
in this behavior change.

This patch also refactors the defect tests into a separate test file,
since they are no longer just parser tests.

12 years ago#14925: email now registers a defect for missing header/body separator.
R David Murray [Mon, 28 May 2012 00:45:01 +0000 (20:45 -0400)]
#14925: email now registers a defect for missing header/body separator.

This patch also deprecates the MalformedHeaderDefect.  My best guess is that
this defect was rendered obsolete by a refactoring of the parser, and the
corresponding defect for the new parser (which this patch introduces) was
overlooked.

12 years agoNull merge of email doc patch already applied to this branch.
R David Murray [Sun, 27 May 2012 21:18:28 +0000 (17:18 -0400)]
Null merge of email doc patch already applied to this branch.

12 years ago#11785: fix the :mod: references in email package submodule titles.
R David Murray [Sun, 27 May 2012 21:17:53 +0000 (17:17 -0400)]
#11785: fix the :mod: references in email package submodule titles.

12 years ago#11785: fix the :mod: references in email package submodule titles.
R David Murray [Sun, 27 May 2012 21:10:36 +0000 (17:10 -0400)]
#11785: fix the :mod: references in email package submodule titles.

Also adds the TOC entry for headerregistry.

12 years agoMake headerregistry fully part of the provisional api.
R David Murray [Sun, 27 May 2012 19:03:38 +0000 (15:03 -0400)]
Make headerregistry fully part of the provisional api.

When I made the checkin of the provisional email policy, I knew that
Address and Group needed to be made accessible from somewhere.  The more
I looked at it, though, the more it became clear that since this is a
provisional API anyway, there's no good reason to hide headerregistry as
a private API.  It was designed to ultimately be part of the public API,
and so it should be part of the provisional API.

This patch fully documents the headerregistry API, and deletes the
abbreviated version of those docs I had added to the provisional policy
docs.

12 years agoChanged executable name computation in test_venv to allow for debug executables.
Vinay Sajip [Sun, 27 May 2012 18:05:36 +0000 (19:05 +0100)]
Changed executable name computation in test_venv to allow for debug executables.

12 years agoRefined venv tests.
Vinay Sajip [Sun, 27 May 2012 17:39:22 +0000 (18:39 +0100)]
Refined venv tests.