Mailman 3 November 2012 - Python-Dev

problems building python2.7
by Chris Withers 10 Nov '12

10 Nov '12

Hi All, I wanted to run the unit tests before checking in the patch for http://bugs.python.org/issue16441, even though it's a trivial change, so I was trying to follow the instructions at: http://docs.python.org/devguide/ I'm on MacOS, so following the "unix" instructions did: ./configure --with-pydebug && make -j2 This appears to have worked, given the end of the output: Python build finished, but the necessary bits to build these modules were not found: _bsddb dl gdbm imageop linuxaudiodev ossaudiodev readline spwd sunaudiodev To find the necessary bits, look in setup.py in detect_modules() for the module's name. running build_scripts creating build/scripts-2.7 copying and adjusting /Users/chris/LocalHG/cpython/Tools/scripts/pydoc -> build/scripts-2.7 copying and adjusting /Users/chris/LocalHG/cpython/Tools/scripts/idle -> build/scripts-2.7 copying and adjusting /Users/chris/LocalHG/cpython/Tools/scripts/2to3 -> build/scripts-2.7 copying and adjusting /Users/chris/LocalHG/cpython/Lib/smtpd.py -> build/scripts-2.7 changing mode of build/scripts-2.7/pydoc from 644 to 755 changing mode of build/scripts-2.7/idle from 644 to 755 changing mode of build/scripts-2.7/2to3 from 644 to 755 changing mode of build/scripts-2.7/smtpd.py from 644 to 755 [65930 refs] However, I can't find the python it's built... I thought I'd be clever and try: buzzkill:cpython chris$ cat build/scripts-2.7/2to3 #!/usr/local/bin/python2.7 ... There is a python there, but it's a symlink put in place around a year ago. So, where should I look for the built python? Okay, so regardless, my change is only to the stdlib, so I thought I'd try the test instructions anyway: /usr/local/bin/python2.7 -m test -j3 gives: /Library/Frameworks/Python.framework/Versions/2.7/Resources/Python.app/Contents/MacOS/Python: No module named test.__main__; 'test' is a package and cannot be directly executed Now, if this had worked, would it have discovered the right gzip tests, or is it going to run whatever came with the python2.7 distro the binary comes from? Chris -- Simplistix - Content Management, Batch Processing & Python Consulting - http://www.simplistix.co.uk

11 17

Summary of Python tracker Issues
by Python tracker 09 Nov '12

09 Nov '12

ACTIVITY SUMMARY (2012-11-02 - 2012-11-09) Python tracker at http://bugs.python.org/ To view or respond to any of the issues listed below, click on the issue. Do NOT respond to this message. Issues counts and deltas: open 3806 ( +2) closed 24418 (+57) total 28224 (+59) Open issues with patches: 1657 Issues opened (42) ================== #13424: Add examples for open’s new opener argument http://bugs.python.org/issue13424 reopened by eric.araujo #16350: zlib.Decompress.decompress() after EOF discards existing value http://bugs.python.org/issue16350 reopened by serhiy.storchaka #16389: re._compiled_typed's lru_cache causes significant degradation http://bugs.python.org/issue16389 opened by pjenvey #16391: add "terminator" ctor argument to logging.StreamHandlers deriv http://bugs.python.org/issue16391 opened by nikicat #16392: import crashes on circular imports in ext modules http://bugs.python.org/issue16392 opened by scoder #16394: Reducing tee() memory footprint http://bugs.python.org/issue16394 opened by cami #16395: Documentation claims that PySequence_Fast returns a tuple, whe http://bugs.python.org/issue16395 opened by sfllaw #16396: Importing ctypes.wintypes on Linux gives a traceback http://bugs.python.org/issue16396 opened by techtonik #16397: UserString doesn't combine nicely with strings http://bugs.python.org/issue16397 opened by Jorge.Cardona #16398: deque.rotate() could be much faster http://bugs.python.org/issue16398 opened by sfllaw #16399: argparse: append action with default list adds to list instead http://bugs.python.org/issue16399 opened by Markus.Amalthea.Magnuson #16400: update default PyPI behavior in docs re: listing versions http://bugs.python.org/issue16400 opened by chris.jerdonek #16401: mention PKG-INFO in the documentation http://bugs.python.org/issue16401 opened by chris.jerdonek #16403: update distutils docs to say that maintainer replaces author http://bugs.python.org/issue16403 opened by chris.jerdonek #16404: Uses of PyLong_FromLong that don't check for errors http://bugs.python.org/issue16404 opened by nedbat #16405: Explain how to set up the whitespace commit hook locally http://bugs.python.org/issue16405 opened by ncoghlan #16406: move the "Uploading Packages" section to distutils/packageinde http://bugs.python.org/issue16406 opened by chris.jerdonek #16408: FD leaks in zipfile http://bugs.python.org/issue16408 opened by Eric.Busboom #16409: urlretrieve regression: first call returns block size as 0 http://bugs.python.org/issue16409 opened by techtonik #16411: zlib.Decompress.decompress() retains pointer to input buffer w http://bugs.python.org/issue16411 opened by nadeem.vawda #16413: Non cross-platform behavior of os.path.split http://bugs.python.org/issue16413 opened by techtonik #16416: Mac OS X: don't use the locale encoding but UTF-8 to encode an http://bugs.python.org/issue16416 opened by haypo #16418: argparse with many choices can generate absurdly long usage me http://bugs.python.org/issue16418 opened by Juraj.Variny #16421: importlib.machinery.ExtensionFileLoader cannot load several mo http://bugs.python.org/issue16421 opened by eudoxos #16422: Decimal constants should be the same for py & c module version http://bugs.python.org/issue16422 opened by yselivanov #16423: urllib data URL recipe http://bugs.python.org/issue16423 opened by panzi #16424: regression: os.path.split('//hostname/foo/bar.txt') http://bugs.python.org/issue16424 opened by techtonik #16425: minidom replaceChild(new_child, old_child) removes new_child e http://bugs.python.org/issue16425 opened by Martin.Kugler #16427: Faster hash implementation http://bugs.python.org/issue16427 opened by serhiy.storchaka #16428: turtle with compound shape doesn't get clicks http://bugs.python.org/issue16428 opened by pythonick #16429: Emit SyntaxWarning for code that risks UnboundLocalError http://bugs.python.org/issue16429 opened by ncoghlan #16434: SocketServer call shutdown in the wrong way http://bugs.python.org/issue16434 opened by Erik.Günther #16436: Link directly to set and frozenset in built-in function docs http://bugs.python.org/issue16436 opened by fossilet #16437: issubclass doc improvement http://bugs.python.org/issue16437 opened by fossilet #16438: Numeric operator predecence confusing http://bugs.python.org/issue16438 opened by fossilet #16442: PATH_MAX vs MAXPATHLEN vs pathconf(..., _PC_PATH_MAX). http://bugs.python.org/issue16442 opened by trent #16443: Add docstrings to regular expression match objects http://bugs.python.org/issue16443 opened by rhettinger #16444: Use support.TESTFN_UNDECODABLE on UNIX http://bugs.python.org/issue16444 opened by haypo #16445: SEGFAULT when deleting Exception.message http://bugs.python.org/issue16445 opened by amaury.forgeotdarc #16446: pdb raises BdbQuit on 'quit' when started with set_trace http://bugs.python.org/issue16446 opened by xdegaye #16447: SEGFAULT when setting type.__name__ http://bugs.python.org/issue16447 opened by amaury.forgeotdarc #16432: Template strings documentation in Python 3 refers to % substit http://bugs.python.org/issue16432 opened by andrewsg Most recent 15 issues with no replies (15) ========================================== #16446: pdb raises BdbQuit on 'quit' when started with set_trace http://bugs.python.org/issue16446 #16442: PATH_MAX vs MAXPATHLEN vs pathconf(..., _PC_PATH_MAX). http://bugs.python.org/issue16442 #16429: Emit SyntaxWarning for code that risks UnboundLocalError http://bugs.python.org/issue16429 #16428: turtle with compound shape doesn't get clicks http://bugs.python.org/issue16428 #16427: Faster hash implementation http://bugs.python.org/issue16427 #16425: minidom replaceChild(new_child, old_child) removes new_child e http://bugs.python.org/issue16425 #16423: urllib data URL recipe http://bugs.python.org/issue16423 #16416: Mac OS X: don't use the locale encoding but UTF-8 to encode an http://bugs.python.org/issue16416 #16406: move the "Uploading Packages" section to distutils/packageinde http://bugs.python.org/issue16406 #16403: update distutils docs to say that maintainer replaces author http://bugs.python.org/issue16403 #16400: update default PyPI behavior in docs re: listing versions http://bugs.python.org/issue16400 #16398: deque.rotate() could be much faster http://bugs.python.org/issue16398 #16386: imp.find_module does not specify registry key it searches on w http://bugs.python.org/issue16386 #16382: Better warnings exception for bad category http://bugs.python.org/issue16382 #16379: SQLite error code not exposed to python http://bugs.python.org/issue16379 Most recent 15 issues waiting for review (15) ============================================= #16447: SEGFAULT when setting type.__name__ http://bugs.python.org/issue16447 #16445: SEGFAULT when deleting Exception.message http://bugs.python.org/issue16445 #16444: Use support.TESTFN_UNDECODABLE on UNIX http://bugs.python.org/issue16444 #16427: Faster hash implementation http://bugs.python.org/issue16427 #16423: urllib data URL recipe http://bugs.python.org/issue16423 #16421: importlib.machinery.ExtensionFileLoader cannot load several mo http://bugs.python.org/issue16421 #16416: Mac OS X: don't use the locale encoding but UTF-8 to encode an http://bugs.python.org/issue16416 #16411: zlib.Decompress.decompress() retains pointer to input buffer w http://bugs.python.org/issue16411 #16408: FD leaks in zipfile http://bugs.python.org/issue16408 #16395: Documentation claims that PySequence_Fast returns a tuple, whe http://bugs.python.org/issue16395 #16392: import crashes on circular imports in ext modules http://bugs.python.org/issue16392 #16389: re._compiled_typed's lru_cache causes significant degradation http://bugs.python.org/issue16389 #16386: imp.find_module does not specify registry key it searches on w http://bugs.python.org/issue16386 #16382: Better warnings exception for bad category http://bugs.python.org/issue16382 #16381: Introduce option to force the interpreter to exit upon MemoryE http://bugs.python.org/issue16381 Top 10 most discussed issues (10) ================================= #14621: Hash function is not randomized properly http://bugs.python.org/issue14621 36 msgs #16218: Python launcher does not support unicode characters http://bugs.python.org/issue16218 32 msgs #16353: add function to os module for getting path to default shell http://bugs.python.org/issue16353 32 msgs #9584: Allow curly brace expansion http://bugs.python.org/issue9584 21 msgs #14794: slice.indices raises OverflowError http://bugs.python.org/issue14794 18 msgs #16392: import crashes on circular imports in ext modules http://bugs.python.org/issue16392 17 msgs #16389: re._compiled_typed's lru_cache causes significant degradation http://bugs.python.org/issue16389 16 msgs #16421: importlib.machinery.ExtensionFileLoader cannot load several mo http://bugs.python.org/issue16421 15 msgs #10030: Patch for zip decryption speedup http://bugs.python.org/issue10030 12 msgs #13349: Non-informative error message in index() and remove() function http://bugs.python.org/issue13349 8 msgs Issues closed (54) ================== #4318: optparse: formatting of help text/descriptions http://bugs.python.org/issue4318 closed by eric.araujo #4711: Wide literals in the table of contents overflow in documentati http://bugs.python.org/issue4711 closed by ezio.melotti #5765: stack overflow evaluating eval("()" * 30000) http://bugs.python.org/issue5765 closed by ncoghlan #5894: Lookup of localised language name by ISO 639 language code and http://bugs.python.org/issue5894 closed by christian.heimes #6709: It's possible to create TryExcept with no handlers http://bugs.python.org/issue6709 closed by benjamin.peterson #7317: Display full tracebacks when an error occurs asynchronously http://bugs.python.org/issue7317 closed by asvetlov #8271: str.decode('utf8', 'replace') -- conformance with Unicode 5.2. http://bugs.python.org/issue8271 closed by ezio.melotti #8401: Strange behavior of bytearray slice assignment http://bugs.python.org/issue8401 closed by ezio.melotti #9109: absolute import cleanups for Python 3 http://bugs.python.org/issue9109 closed by akuchling #10358: Doc styles for print should only use dark colors http://bugs.python.org/issue10358 closed by ezio.melotti #10385: Mark up "subprocess" as module in its doc http://bugs.python.org/issue10385 closed by ezio.melotti #11166: No exit when daemon thread is running. http://bugs.python.org/issue11166 closed by akuchling #11313: Speed up default encode()/decode() http://bugs.python.org/issue11313 closed by ezio.melotti #11481: The copy module already uses copyreg http://bugs.python.org/issue11481 closed by ezio.melotti #11631: Python 2.7.1 64bit, Win7 64bit problem to read and write with http://bugs.python.org/issue11631 closed by christian.heimes #11842: slice.indices with negative step and default stop http://bugs.python.org/issue11842 closed by mark.dickinson #12759: "(?P=)" input for Tools/scripts/redemo.py raises unnhandled ex http://bugs.python.org/issue12759 closed by ezio.melotti #13301: the script Tools/i18n/msgfmt.py allows arbitrary code executio http://bugs.python.org/issue13301 closed by ezio.melotti #15001: segmentation fault with del sys.modules['__main__'] http://bugs.python.org/issue15001 closed by hynek #15076: Sometimes couldn't import os, shown 'import site' failed, use http://bugs.python.org/issue15076 closed by akuchling #15165: test_email: failure on Windows http://bugs.python.org/issue15165 closed by python-dev #15641: Clean up importlib for Python 3.4 http://bugs.python.org/issue15641 closed by asvetlov #15814: memoryview: equality-hash invariant http://bugs.python.org/issue15814 closed by skrah #15837: Added test to test_random.py testing Random.shuffle http://bugs.python.org/issue15837 closed by pitrou #16048: Tutorial-classes-remarks: replace paragraph http://bugs.python.org/issue16048 closed by ezio.melotti #16152: Trailing whitespace makes tokenize.generate_tokens pathologica http://bugs.python.org/issue16152 closed by ezio.melotti #16183: ZipExtFile object close without file handle closed http://bugs.python.org/issue16183 closed by python-dev #16284: concurrent.futures ThreadPoolExecutor keeps unnecessary refere http://bugs.python.org/issue16284 closed by asvetlov #16304: re: Match Objects always have a boolean value of True http://bugs.python.org/issue16304 closed by ezio.melotti #16309: "PYTHONPATH=" different from no PYTHONPATH at all http://bugs.python.org/issue16309 closed by asvetlov #16311: Use _PyUnicodeWriter API in text decoders http://bugs.python.org/issue16311 closed by python-dev #16336: Check input in surrogatepass error handler http://bugs.python.org/issue16336 closed by ezio.melotti #16338: pysnmp/asyncore - timeout ineffective? http://bugs.python.org/issue16338 closed by neologix #16354: Remember python version choice on docs.python.org http://bugs.python.org/issue16354 closed by georg.brandl #16385: evaluating literal dict with repeated keys gives no warnings/e http://bugs.python.org/issue16385 closed by benjamin.peterson #16390: re compilation slow in Python 3.3 due to functools.lru_cache o http://bugs.python.org/issue16390 closed by pjenvey #16393: typo in pkgutil.py http://bugs.python.org/issue16393 closed by ezio.melotti #16402: range slicing swallows exceptions from __index__ methods. http://bugs.python.org/issue16402 closed by mark.dickinson #16407: Python 2.7.3: test_sys fails when building under 10.7.5 with d http://bugs.python.org/issue16407 closed by ned.deily #16410: rewrite winsound with ctypes http://bugs.python.org/issue16410 closed by loewis #16412: test_shutil: 2 failures related to symlinks http://bugs.python.org/issue16412 closed by pitrou #16414: Add support.NONASCII to test non-ASCII characters http://bugs.python.org/issue16414 closed by haypo #16415: eventlet.monkey_patch() breaks subprocess.Popen on Windows http://bugs.python.org/issue16415 closed by r.david.murray #16417: Integer converted in tuple without request http://bugs.python.org/issue16417 closed by r.david.murray #16419: email.message._headers is a list http://bugs.python.org/issue16419 closed by ezio.melotti #16420: PEP 249 (DB-API 2.0) converted to reStructuredText http://bugs.python.org/issue16420 closed by lemburg #16426: RoundUp signature removal test (please ignore) http://bugs.python.org/issue16426 closed by lemburg #16430: re.match blocking and taking 100% CPU http://bugs.python.org/issue16430 closed by serhiy.storchaka #16431: CDecimal disregards subclass passed into __new__ when value ar http://bugs.python.org/issue16431 closed by skrah #16433: unittest.TestCase.assertNotEqual has incorrect docstring. http://bugs.python.org/issue16433 closed by ezio.melotti #16435: Python 3 doc link to Python 2 FAQ http://bugs.python.org/issue16435 closed by asvetlov #16439: Code not collapsed correctly http://bugs.python.org/issue16439 closed by asvetlov #16440: Exception type http://bugs.python.org/issue16440 closed by ezio.melotti #16441: range usage in gzip module leads to excessive memory usage. http://bugs.python.org/issue16441 closed by cjw296

1 0

Re: [Python-Dev] cpython: issue9584: Add {} list expansion to glob. Original patch by Mathieu Bridon
by Georg Brandl 09 Nov '12

09 Nov '12

On 11/06/2012 02:56 PM, tim.golden wrote: > http://hg.python.org/cpython/rev/dafca4714298 > changeset: 80273:dafca4714298 > user: Tim Golden <mail(a)timgolden.me.uk> > date: Tue Nov 06 13:50:42 2012 +0000 > summary: > issue9584: Add {} list expansion to glob. Original patch by Mathieu Bridon > > files: > Doc/library/glob.rst | 11 ++-- > Lib/glob.py | 65 ++++++++++++++++++++++-------- > Lib/test/test_glob.py | 64 +++++++++++++++++++++++++++++- > Misc/NEWS | 3 + > 4 files changed, 118 insertions(+), 25 deletions(-) > > > diff --git a/Doc/library/glob.rst b/Doc/library/glob.rst > --- a/Doc/library/glob.rst > +++ b/Doc/library/glob.rst > @@ -13,10 +13,10 @@ > > The :mod:`glob` module finds all the pathnames matching a specified pattern > according to the rules used by the Unix shell. No tilde expansion is done, but > -``*``, ``?``, and character ranges expressed with ``[]`` will be correctly > -matched. This is done by using the :func:`os.listdir` and > -:func:`fnmatch.fnmatch` functions in concert, and not by actually invoking a > -subshell. (For tilde and shell variable expansion, use > +``*``, ``?``, character ranges expressed with ``[]`` and list of options > +expressed with ``{}`` will be correctly matched. This is done by using the > +:func:`os.listdir` and :func:`fnmatch.fnmatch` functions in concert, and not by > +actually invoking a subshell. (For tilde and shell variable expansion, use > :func:`os.path.expanduser` and :func:`os.path.expandvars`.) Needs a versionchanged. In any case, brace expansion is not part of globbing (see the bash or zsh manuals) because it does not generate valid file names, and it is a non-POSIX expansion of some shells. Are you sure it should be put into the glob module? (Not speaking of the backward incompatibility it creates.) Georg

2 2

Detecting tp_compare / tp_richcompare from Python
by Tres Seaver 09 Nov '12

09 Nov '12

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 While porting the BTrees package (split out from ZODB) to Python3., my first step is to get the pure-Python reference implementation working: in order to do that, I need a way to check that objects used as keys are *not* using the comparison semantics inherited from the base 'object' class, because those semantics rely on properties which are not stable across the lifetime of the (persisted / restored) key. The existing C code does something like:: static int check_argument_cmp(PyObject *arg) { if (arg->ob_type->tp_richcompare == NULL && arg->ob_type->tp_compare == ((PyTypeObject *)object_)->ob_type->tp_compare) { PyErr_SetString(PyExc_TypeError, "Object has default comparison"); return 0; } return 1; } Unless I'm mistaken, there is no way to do the equivalent from pure Python.. I tried a couple of approximations which relied on non-API attributes (I"m looking at out, methodwrapper.__objclass__), but without much success (and I need the code to work on PyPy / Jython, too). Am I missing something? Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver(a)palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iEYEARECAAYFAlCb21gACgkQ+gerLs4ltQ4jBACfV0lQaQ2eW2vhAtWunLUsPQWM esEAoMYeeQvlJVnckaBg4HM19LoxPIWB =+d+0 -----END PGP SIGNATURE-----

2 5

Re: [Python-Dev] cpython: Issue #16218: skip test if filesystem doesn't support required encoding
by Antoine Pitrou 08 Nov '12

08 Nov '12

On Sat, 3 Nov 2012 13:37:48 +0100 (CET) andrew.svetlov <python-checkins(a)python.org> wrote: > http://hg.python.org/cpython/rev/95d1adf144ee > changeset: 80187:95d1adf144ee > user: Andrew Svetlov <andrew.svetlov(a)gmail.com> > date: Sat Nov 03 14:37:37 2012 +0200 > summary: > Issue #16218: skip test if filesystem doesn't support required encoding > > files: > Lib/test/test_cmd_line_script.py | 7 ++++++- > 1 files changed, 6 insertions(+), 1 deletions(-) > > > diff --git a/Lib/test/test_cmd_line_script.py b/Lib/test/test_cmd_line_script.py > --- a/Lib/test/test_cmd_line_script.py > +++ b/Lib/test/test_cmd_line_script.py > @@ -366,7 +366,12 @@ > def test_non_utf8(self): > # Issue #16218 > with temp_dir() as script_dir: > - script_basename = '\udcf1\udcea\udcf0\udce8\udcef\udcf2' > + script_basename = '\u0441\u043a\u0440\u0438\u043f\u0442' Why exactly did you change the tested name here? Regards Antoine.

5 6

chained assignment weirdity
by Chris Withers 08 Nov '12

08 Nov '12

Hi All, I bumped into this using Michael Foord's Mock library. It feels like a bug to me, but thought I'd ask here before logging one in the tracker in case people know that we won't be able to fix it: On 05/11/2012 13:43, Michael Foord wrote: > >>>>>> class Foo(object): >>> ... def __setattr__(s, k, v): >>> ... print k >>> ... >>>>>> f = Foo() >>>>>> >>>>>> f.g = f.x = 3 >>> g >>> x >> >> Hmm, that's definitely counter-intuitive. Worth raising on python-dev? > > Well, I guess "it's always been that way", so changing it would be backwards incompatible. Feel free to raise it though. :-) Here's the actual problem I had: >>>>>>> mock = Mock() >>>>>>> inst = mock.Popen.return_value = mock.Popen_instance = Mock(spec=Popen) >>> >>> >>> Here the *return_value* is set first. So when mock.Popen_instance is assigned it already has a parent! (Setting a mock as the return value of another mock also establishes the child / parent relationship.) >>> >>> This is why calls to the instance are then not recorded in "method_calls". cheers, Chris -- Simplistix - Content Management, Batch Processing & Python Consulting - http://www.simplistix.co.uk

14 31

urlretrieve regression no.2, now in Python 3.3
by anatoly techtonik 08 Nov '12

08 Nov '12

urlretrieve has a callback parameter, which takes function with the following prototype: def callback(block_number, block_size, total_size): pass Where block_size was constant and block_size*block_number gave an exact number of transferred bytes. Recent change in Python 3.3 changed the semantic of block_size and broke my `wget` package. http://bugs.python.org/issue16409 Can anybody tell me if I can wait for the fix in 3.3.1 or should implement a long-term workaround myself? -- anatoly t.

3 2

Fwd: [issue16424] regression: os.path.split('//hostname/foo/bar.txt')
by anatoly techtonik 07 Nov '12

07 Nov '12

Forwarding to python-dev. Does everybody agree that the following behavior is not a regression, but a feature of os.path.split()? Python 3: >>> import os.path as osp >>> osp.split('//hostname/foo/') ('//hostname/foo/', '') Python 2: >>> osp.split('//hostname/foo/') ('//hostname/foo', '') But Python 3 again: >>> osp.split('//hostname/foo/bar/') ('//hostname/foo/bar', '') http://bugs.python.org/issue16424

1 0

回复： Python-Dev Digest, Vol 112, Issue 8
by 邓九祥 07 Nov '12

07 Nov '12

I hava some question about Object which "inital" need ,but return "object" in mongo shell. So BSONElement initial = p["initial"]; if ( initial.type() != Object ) { errmsg = "initial has to be an object"; return false; } "initial.type() != Object " will be found ,so "group" command will not run forever. What can'i do Mybe this is my fase, please give me some advise.

2 1

Re: [Python-Dev] cpython (3.3): Add examples for opener argument of open (#13424).
by Georg Brandl 06 Nov '12

06 Nov '12

Am 03.11.2012 22:10, schrieb eric.araujo: > http://hg.python.org/cpython/rev/17b094c08600 > changeset: 80219:17b094c08600 > branch: 3.3 > parent: 80214:e6d0951f412a > user: Éric Araujo <merwok(a)netwok.org> > date: Sat Nov 03 17:06:52 2012 -0400 > summary: > Add examples for opener argument of open (#13424). > > Patch by Guillaume Pratte. > > files: > Doc/library/functions.rst | 30 +++++++++++++++++++++++++++ > Doc/library/io.rst | 3 ++ > Misc/ACKS | 1 + > 3 files changed, 34 insertions(+), 0 deletions(-) > > > diff --git a/Doc/library/functions.rst b/Doc/library/functions.rst > --- a/Doc/library/functions.rst > +++ b/Doc/library/functions.rst > @@ -935,6 +935,36 @@ > :mod:`os.open` as *opener* results in functionality similar to passing > ``None``). > > + The following example is an alternative implementation for opening files > + for exclusive writing. If we did not have support for the ``'x'`` mode, > + we could implement it with this opener:: > + > + >>> import os > + >>> def open_exclusive(path, mode): > + ... return os.open(path, mode | os.O_CREAT | os.O_EXCL) > + ... > + >>> filename = 'spam.txt' > + >>> fp = open(filename, 'w', opener=open_exclusive) > + >>> fp2 = open(filename, 'w', opener=open_exclusive) > + Traceback (most recent call last): > + ... > + FileExistsError: [Errno 17] File exists: 'spam.txt' > + > + This other example uses the :ref:`dir_fd` parameter of the > + :func:`os.open` function to open a file relative to a given directory:: Please heed your Sphinx warnings: the :ref:`dir_fd` needs a link caption, since it can't autogenerate one (the dir_fd anchor does not point to a heading). Georg

2 1