Mailman 3 July 2013 - Python-checkins

cpython (2.7): Closes #18312: 'make distclean' no longer deletes files in dot-directories.
by eric.smith July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/5896f887a93a changeset: 84431:5896f887a93a branch: 2.7 parent: 84428:20b59fec1123 user: Eric V. Smith <eric(a)trueblade.com> date: Tue Jul 02 09:02:03 2013 -0400 summary: Closes #18312: 'make distclean' no longer deletes files in dot-directories. files: Makefile.pre.in | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/Makefile.pre.in b/Makefile.pre.in --- a/Makefile.pre.in +++ b/Makefile.pre.in @@ -1295,11 +1295,11 @@ Modules/ld_so_aix Modules/python.exp Misc/python.pc -rm -f python*-gdb.py -rm -f pybuilddir.txt - find $(srcdir) '(' -name '*.fdc' -o -name '*~' \ - -o -name '[@,#]*' -o -name '*.old' \ - -o -name '*.orig' -o -name '*.rej' \ - -o -name '*.bak' ')' \ - -exec rm -f {} ';' + find $(srcdir)/[a-zA-Z]* '(' -name '*.fdc' -o -name '*~' \ + -o -name '[@,#]*' -o -name '*.old' \ + -o -name '*.orig' -o -name '*.rej' \ + -o -name '*.bak' ')' \ + -exec rm -f {} ';' # Check for smelly exported symbols (not starting with Py/_Py) smelly: all -- Repository URL: http://hg.python.org/cpython

1 0

devinabox: Tweak a few things
by nick.coghlan July 2, 2013

July 2, 2013

http://hg.python.org/devinabox/rev/53a1f44a6219 changeset: 48:53a1f44a6219 user: Nick Coghlan <ncoghlan(a)gmail.com> date: Tue Jul 02 22:45:52 2013 +1000 summary: Tweak a few things files: .hgignore | 2 ++ README | 22 +++++++++++++++------- full_coverage.py | 13 ++++++++++--- 3 files changed, 27 insertions(+), 10 deletions(-) diff --git a/.hgignore b/.hgignore --- a/.hgignore +++ b/.hgignore @@ -5,6 +5,8 @@ cpython devguide peps +# Unversioned downloads +Visual C++ Express/vc_web.exe # Generated coverage_report original_coverage_report diff --git a/README b/README --- a/README +++ b/README @@ -26,6 +26,10 @@ get updates, sparing the probably taxed internet connection at the sprint from doing complete repository cloning. +If recreating from an old checkout ``hg purge --all`` in the individual +clones is a handy way to ensure old build artifacts have been removed. +You will need to enable the purge extension in ``~/.hgrc``. + Mercurial --------- @@ -104,15 +108,16 @@ 01. Build the CPython repository 02. Clone the repository from https://bitbucket.org/ned/coveragepy -03. Download Distribute from https://pypi.python.org/pypi/distribute -04. Unpack and build Distribute with Python 3 in the coveragepy repository -05. Symlink the ``distribute-N.N.N/build/lib/setuptools`` directory into - ``coveragepy`` -06. Symlink ``distribute-N.N.N/build/lib/pkg_resources.py`` into ``coveragepy`` +03. Clone setuptools from https://bitbucket.org/pypa/setuptools/ +04. Run ``python3 setup.py build`` in the ``setuptools`` directory +05. Run ``ln -s ../setuptools/build/lib/setuptools`` in the ``coveragepy`` + directory +06. Run ``ln -s ../setuptools/build/lib/pkg_resources.py`` in the + ``coveragepy`` directory 07. Run ``./cpython/python full_coverage.py build`` 08. Run ``./cpython/python full_coverage.py run`` 09. Run ``./cpython/python full_coverage.py html original_coverage_report`` -10. ``make distclean`` the CPython repository +10. Run ``hg purge --all`` in the CPython repository All these steps will generate a complete coverage report for the standard @@ -150,9 +155,12 @@ just simplifies this by having a single command subsume both the configure and build steps. It also uses reasonable defaults (e.g. all cores on the CPU). +(You may need to cd into the CPython directory and run ``make`` to get the +extension modules to build) + ``index.html`` -------------- An HTML file with links to the various pieces of documentation you built -previously and the helper scripts. \ No newline at end of file +previously and the helper scripts. diff --git a/full_coverage.py b/full_coverage.py --- a/full_coverage.py +++ b/full_coverage.py @@ -29,6 +29,13 @@ yield os.chdir(original_directory) +def make_setup_cmd(*args): + # Silence a spurious warning from setuptools + # See https://bitbucket.org/pypa/setuptools/issue/29/ + cmd = [sys.executable, '-W', 'ignore::UserWarning:distutils.dist', + 'setup.py'] + cmd.extend(args) + return cmd def build(args): """Build coverage.py's C-based tracer. @@ -43,11 +50,11 @@ os.unlink(tracer_path) except FileNotFoundError: pass - subprocess.check_call([sys.executable, 'setup.py', 'clean']) + subprocess.check_call(make_setup_cmd('clean')) env = os.environ.copy() env['CPPFLAGS'] = '-I {} -I {}'.format(CPYTHON, os.path.join(CPYTHON, 'Include')) - command = [sys.executable, 'setup.py', 'build_ext', '--inplace'] + command = make_setup_cmd('build_ext', '--inplace') process = subprocess.Popen(command, env=env) process.wait() @@ -124,4 +131,4 @@ help_args = ['-h'] if args.command: help_args.insert(0, args.command) - parser.parse_args(help_args) \ No newline at end of file + parser.parse_args(help_args) -- Repository URL: http://hg.python.org/devinabox

1 0

cpython (merge 3.3 -> default): Issue #17261: Ensure multiprocessing's proxies use proper address.
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/7387b884f833 changeset: 84430:7387b884f833 parent: 84427:d0b48efac9de parent: 84429:be4b9e677187 user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 13:38:58 2013 +0100 summary: Issue #17261: Ensure multiprocessing's proxies use proper address. files: Lib/multiprocessing/managers.py | 1 + Misc/NEWS | 2 ++ 2 files changed, 3 insertions(+), 0 deletions(-) diff --git a/Lib/multiprocessing/managers.py b/Lib/multiprocessing/managers.py --- a/Lib/multiprocessing/managers.py +++ b/Lib/multiprocessing/managers.py @@ -731,6 +731,7 @@ elif kind == '#PROXY': exposed, token = result proxytype = self._manager._registry[token.typeid][-1] + token.address = self._token.address proxy = proxytype( token, self._serializer, manager=self._manager, authkey=self._authkey, exposed=exposed diff --git a/Misc/NEWS b/Misc/NEWS --- a/Misc/NEWS +++ b/Misc/NEWS @@ -138,6 +138,8 @@ Library ------- +- Issue #17261: Ensure multiprocessing's proxies use proper address. + - Issue #18343: faulthandler.register() now keeps the previous signal handler when the function is called twice, so faulthandler.unregister() restores correctly the original signal handler. -- Repository URL: http://hg.python.org/cpython

1 0

cpython (3.3): Issue #17261: Ensure multiprocessing's proxies use proper address.
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/be4b9e677187 changeset: 84429:be4b9e677187 branch: 3.3 parent: 84426:9746f217a270 user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 13:37:43 2013 +0100 summary: Issue #17261: Ensure multiprocessing's proxies use proper address. files: Lib/multiprocessing/managers.py | 1 + Misc/NEWS | 2 ++ 2 files changed, 3 insertions(+), 0 deletions(-) diff --git a/Lib/multiprocessing/managers.py b/Lib/multiprocessing/managers.py --- a/Lib/multiprocessing/managers.py +++ b/Lib/multiprocessing/managers.py @@ -731,6 +731,7 @@ elif kind == '#PROXY': exposed, token = result proxytype = self._manager._registry[token.typeid][-1] + token.address = self._token.address proxy = proxytype( token, self._serializer, manager=self._manager, authkey=self._authkey, exposed=exposed diff --git a/Misc/NEWS b/Misc/NEWS --- a/Misc/NEWS +++ b/Misc/NEWS @@ -41,6 +41,8 @@ Library ------- +- Issue #17261: Ensure multiprocessing's proxies use proper address. + - Issue #18343: faulthandler.register() now keeps the previous signal handler when the function is called twice, so faulthandler.unregister() restores correctly the original signal handler. -- Repository URL: http://hg.python.org/cpython

1 0

cpython (2.7): Issue #17261: Ensure multiprocessing's proxies use proper address.
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/20b59fec1123 changeset: 84428:20b59fec1123 branch: 2.7 parent: 84425:f35401dba89f user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 13:31:43 2013 +0100 summary: Issue #17261: Ensure multiprocessing's proxies use proper address. files: Lib/multiprocessing/managers.py | 1 + Misc/NEWS | 2 ++ 2 files changed, 3 insertions(+), 0 deletions(-) diff --git a/Lib/multiprocessing/managers.py b/Lib/multiprocessing/managers.py --- a/Lib/multiprocessing/managers.py +++ b/Lib/multiprocessing/managers.py @@ -763,6 +763,7 @@ elif kind == '#PROXY': exposed, token = result proxytype = self._manager._registry[token.typeid][-1] + token.address = self._token.address proxy = proxytype( token, self._serializer, manager=self._manager, authkey=self._authkey, exposed=exposed diff --git a/Misc/NEWS b/Misc/NEWS --- a/Misc/NEWS +++ b/Misc/NEWS @@ -24,6 +24,8 @@ Library ------- +- Issue #17261: Ensure multiprocessing's proxies use proper address. + - Issue #17097: Make multiprocessing ignore EINTR. - Issue #18155: The csv module now correctly handles csv files that use -- Repository URL: http://hg.python.org/cpython

1 0

cpython (merge 3.3 -> default): Issue #14206: Clarify docs for Queue.join_cancel_thread().
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/d0b48efac9de changeset: 84427:d0b48efac9de parent: 84424:7ccf3d36ad13 parent: 84426:9746f217a270 user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 13:01:31 2013 +0100 summary: Issue #14206: Clarify docs for Queue.join_cancel_thread(). files: Doc/library/multiprocessing.rst | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/Doc/library/multiprocessing.rst b/Doc/library/multiprocessing.rst --- a/Doc/library/multiprocessing.rst +++ b/Doc/library/multiprocessing.rst @@ -659,6 +659,13 @@ the background thread from being joined automatically when the process exits -- see :meth:`join_thread`. + A better name for this method might be + ``allow_exit_without_flush()``. It is likely to cause enqueued + data to lost, and you almost certainly will not need to use it. + It is really only there if you need the current process to exit + immediately without waiting to flush enqueued data to the + underlying pipe, and you don't care about lost data. + .. class:: SimpleQueue() -- Repository URL: http://hg.python.org/cpython

1 0

cpython (3.3): Issue #14206: Clarify docs for Queue.join_cancel_thread().
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/9746f217a270 changeset: 84426:9746f217a270 branch: 3.3 parent: 84423:57fe80fda9be user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 12:59:55 2013 +0100 summary: Issue #14206: Clarify docs for Queue.join_cancel_thread(). files: Doc/library/multiprocessing.rst | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/Doc/library/multiprocessing.rst b/Doc/library/multiprocessing.rst --- a/Doc/library/multiprocessing.rst +++ b/Doc/library/multiprocessing.rst @@ -659,6 +659,13 @@ the background thread from being joined automatically when the process exits -- see :meth:`join_thread`. + A better name for this method might be + ``allow_exit_without_flush()``. It is likely to cause enqueued + data to lost, and you almost certainly will not need to use it. + It is really only there if you need the current process to exit + immediately without waiting to flush enqueued data to the + underlying pipe, and you don't care about lost data. + .. class:: SimpleQueue() -- Repository URL: http://hg.python.org/cpython

1 0

cpython (2.7): Issue #14206: Clarify docs for Queue.join_cancel_thread().
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/f35401dba89f changeset: 84425:f35401dba89f branch: 2.7 parent: 84422:389788ba6bcb user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 12:58:21 2013 +0100 summary: Issue #14206: Clarify docs for Queue.join_cancel_thread(). files: Doc/library/multiprocessing.rst | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/Doc/library/multiprocessing.rst b/Doc/library/multiprocessing.rst --- a/Doc/library/multiprocessing.rst +++ b/Doc/library/multiprocessing.rst @@ -630,6 +630,13 @@ the background thread from being joined automatically when the process exits -- see :meth:`join_thread`. + A better name for this method might be + ``allow_exit_without_flush()``. It is likely to cause enqueued + data to lost, and you almost certainly will not need to use it. + It is really only there if you need the current process to exit + immediately without waiting to flush enqueued data to the + underlying pipe, and you don't care about lost data. + .. class:: multiprocessing.queues.SimpleQueue() -- Repository URL: http://hg.python.org/cpython

1 0

cpython (merge 3.3 -> default): Issue #17273: Clarify that pool methods can only be used by parent process.
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/7ccf3d36ad13 changeset: 84424:7ccf3d36ad13 parent: 84421:74b7ff20e0e4 parent: 84423:57fe80fda9be user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 12:41:00 2013 +0100 summary: Issue #17273: Clarify that pool methods can only be used by parent process. files: Doc/library/multiprocessing.rst | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/Doc/library/multiprocessing.rst b/Doc/library/multiprocessing.rst --- a/Doc/library/multiprocessing.rst +++ b/Doc/library/multiprocessing.rst @@ -284,6 +284,9 @@ print(result.get(timeout=1)) # prints "100" unless your computer is *very* slow print(pool.map(f, range(10))) # prints "[0, 1, 4,..., 81]" +Note that the methods of a pool should only ever be used by the +process which created it. + Reference --------- @@ -1668,6 +1671,9 @@ *initializer* is not ``None`` then each worker process will call ``initializer(*initargs)`` when it starts. + Note that the methods of the pool object should only be called by + the process which created the pool. + .. versionadded:: 3.2 *maxtasksperchild* is the number of tasks a worker process can complete before it will exit and be replaced with a fresh worker process, to enable -- Repository URL: http://hg.python.org/cpython

1 0

cpython (3.3): Issue #17273: Clarify that pool methods can only be used by parent process.
by richard.oudkerk July 2, 2013

July 2, 2013

http://hg.python.org/cpython/rev/57fe80fda9be changeset: 84423:57fe80fda9be branch: 3.3 parent: 84420:229dde749ed6 user: Richard Oudkerk <shibturn(a)gmail.com> date: Tue Jul 02 12:32:00 2013 +0100 summary: Issue #17273: Clarify that pool methods can only be used by parent process. files: Doc/library/multiprocessing.rst | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/Doc/library/multiprocessing.rst b/Doc/library/multiprocessing.rst --- a/Doc/library/multiprocessing.rst +++ b/Doc/library/multiprocessing.rst @@ -284,6 +284,9 @@ print(result.get(timeout=1)) # prints "100" unless your computer is *very* slow print(pool.map(f, range(10))) # prints "[0, 1, 4,..., 81]" +Note that the methods of a pool should only ever be used by the +process which created it. + Reference --------- @@ -1665,6 +1668,9 @@ *initializer* is not ``None`` then each worker process will call ``initializer(*initargs)`` when it starts. + Note that the methods of the pool object should only be called by + the process which created the pool. + .. versionadded:: 3.2 *maxtasksperchild* is the number of tasks a worker process can complete before it will exit and be replaced with a fresh worker process, to enable -- Repository URL: http://hg.python.org/cpython

1 0