Mailman 3 March 2011 - Python-Dev

hg merge flow
by "Martin v. Löwis" 05 Mar '11

05 Mar '11

How should patches be applied to the cpython repository if they land in more than one branch? More specifically, assuming I want to patch all of 2.7, 3.1, 3.2, and default - how do I apply the patches and how do I merge? Regards, Martin

2 1

Mercurial conversion repositories
by Antoine Pitrou 05 Mar '11

05 Mar '11

Hello, Georg and I have been working on converting the SVN repository to Mercurial. We can now present you a test repository (actually, two). CPython repository: http://hg.python.org/cpython/ ------------------ This is the main conversion repository. It contains all history of trunk and py3k (since 1990!) up to now, including all maintenance branches starting from 2.0 up to 3.2. If you are a core developer, get your local clone of the repository using: $ hg clone ssh://hg@hg.python.org/cpython (this uses the same SSH key as your Subversion access; for more information about Mercurial and SSH keys, see the converted development FAQ: http://potrou.net/hgdevguide/faq.html#faq ) If you are not a core developer: $ hg clone http://hg.python.org/cpython Your clone will contain the following branches: $ hg branches default 68026:f12ef116dd10 3.2 68025:cef92ee1a323 2.7 68010:8174d00d0797 3.1 67955:5be8b695ea86 2.6 67287:5e26a860eded 2.5 65464:e4ecac76e499 trunk 62750:800f6c92c3ed 3.0 60075:1d05144224fe 2.4 58552:df72cac1899e 2.3 45731:a3d9a9730743 2.2 40444:d55ddc8c8501 2.1 30171:06fcccf6eca8 2.0 18214:dc0dfc9565cd The branch "default" is the current py3k branch from SVN. The branch "trunk" represents SVN trunk history until 2.7 was branched for maintenance. The full list of tags is too long to print here, but you can get it using: $ hg tags The size of the repository on-disk is (depending on your filesystem): $ du -hs .hg 176M .hg (the size of the network transfer is estimated to be around 80MB) You can commit and even push to this repository (the latter if you are a core developer). Since this is a test repository, whatever you push will be discarded when we do the final conversion. CPython with extra history: http://hg.python.org/cpythonextrahist/ -------------------------- This repository is bigger, and has a much more complicated topology. It is a superset of the other repository, and contains the totality of the branches from the SVN repository (it has more than 450 repository heads, of which 87 non-closed). It also weighs quite a bit more: $ du -hs .hg 248M .hg This repository is unnecessary for development work, since even for history-digging purposes the normal repository has the necessary information. This repository is only to preserve historical record of some of the non-trunk development work from the SVN repository (such as orphaned/deleted feature branches). Development guide: http://potrou.net/hgdevguide/ ----------------- This is the development guide adapted for Mercurial. You can get its sources from the branch "hg_transition" in http://hg.python.org/devguide/. Regards Antoine.

23 97

contributors survey?
by Antoine Pitrou 05 Mar '11

05 Mar '11

Hello, In http://mail.python.org/pipermail/python-committers/2011-February/001340.html, I was asking whether it would be useful to make a survey of past contributors, as in: First, we did a survey of all our past developers who had left the project, asking them why they had left. This was just a free-form survey, allowing people to answer any way they wanted. (from the quoted article in the thread linked above) Since I didn't get any answer, I wonder if the idea simply got overlooked, or if there's no need at all. Regards Antoine.

15 29

rXXX links in the bug tracker after the migration to Mercurial
by Victor Stinner 05 Mar '11

05 Mar '11

Hi, Does the bug tracker will continue to support rXXXXX links after the migration to Mercurial? Victor

6 11

Re: [Python-Dev] [Python-checkins] pymigr: The peps repo is the only other one that seems relevant
by "Martin v. Löwis" 05 Mar '11

05 Mar '11

> The peps repo is the only other one that seems relevant Actually, the stackless people also requested that their repository gets converted. Regards, Martin

1 0

Rietveld or Review Board use?
by skip＠pobox.com 05 Mar '11

05 Mar '11

Is Rietveld or Review Board being used within the Python core development community? I looked at the dev guide but didn't see anything obvious about code reviews. I don't see how to search the Rietveld instance at codereview.appspot.com looking just for Python core review requests. My aim here is to provide some examples to a colleague at work. I wanted to give him some examples related to code with which I'm familiar. Thx, Skip

7 7

Summary of Python tracker Issues
by Python tracker 04 Mar '11

04 Mar '11

ACTIVITY SUMMARY (2011-02-25 - 2011-03-04) 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 2691 ( +9) closed 20487 (+65) total 23178 (+74) Open issues with patches: 1143 Issues opened (50) ================== #11258: ctypes: Speed up find_library() on Linux by 500% http://bugs.python.org/issue11258 reopened by pitrou #11324: ConfigParser(interpolation=None) doesn't work http://bugs.python.org/issue11324 opened by tbrink #11329: PyEval_InitThreads() not safe before Py_Initialize() http://bugs.python.org/issue11329 opened by Juraj.Ivancic #11332: Increase logging/__init__.py coverage to 97% http://bugs.python.org/issue11332 opened by drakeol #11333: Add empty __slots__ to collections.abc abstract base classes http://bugs.python.org/issue11333 opened by durban #11335: Memory leak after key function failure in sort http://bugs.python.org/issue11335 opened by stutzbach #11337: Nothing refers to footnote [1] on page "6. Simple Statements" http://bugs.python.org/issue11337 opened by MLModel #11338: No list of Python hg repositories http://bugs.python.org/issue11338 opened by blokeley #11339: annotation for class being defined http://bugs.python.org/issue11339 opened by andyharrington #11342: ResourceWarning: unclosed file <_io.TextIOWrapper http://bugs.python.org/issue11342 opened by cjejuni2000 #11343: Make errors due to full parser stack identifiable http://bugs.python.org/issue11343 opened by Trundle #11344: Add height argument to os.path.dirname() http://bugs.python.org/issue11344 opened by blokeley #11346: Generator object should be mentioned in gc module document http://bugs.python.org/issue11346 opened by ishimoto #11347: libpython3.so: Broken soname and linking http://bugs.python.org/issue11347 opened by Arfrever #11349: _pickle should implement the module finalisation protocol http://bugs.python.org/issue11349 opened by ncoghlan #11350: __setitem__()'s problem of dbm.dumb object pointed out by comm http://bugs.python.org/issue11350 opened by ysj.ray #11352: Bug in cgi module doc http://bugs.python.org/issue11352 opened by quentel #11354: argparse: nargs could accept range of options count http://bugs.python.org/issue11354 opened by wm #11355: os.mkdir() and os.mkdirat() don't apply SUID/SGID permissions http://bugs.python.org/issue11355 opened by Arfrever #11357: Add support for PEP 381 -- Mirror Authenticity http://bugs.python.org/issue11357 opened by kelseyhightower #11361: suggestion for os.kill(pid,CTRL_C_EVENT) http://bugs.python.org/issue11361 opened by iigor #11362: image/webp missing from mimetypes.py http://bugs.python.org/issue11362 opened by Richard.Rabbat #11363: Curses - add missing functions to doc http://bugs.python.org/issue11363 opened by sandro.tosi #11364: Move from distutils.sysconfig to sysconfig in test_osx_env http://bugs.python.org/issue11364 opened by eric.araujo #11365: Integrate Buildroot patches (cross-compilation) http://bugs.python.org/issue11365 opened by haypo #11366: xml.etree.ElementTree.Element: docs should mention immediate s http://bugs.python.org/issue11366 opened by techtonik #11367: xml.etree.ElementTree.find(all): docs are wrong http://bugs.python.org/issue11367 opened by techtonik #11368: Document why xml.etree.ElementTree.Element has no reference to http://bugs.python.org/issue11368 opened by techtonik #11369: Add caching for the isEnabledFor() computation http://bugs.python.org/issue11369 opened by William.Hart #11370: Fix distutils to carry configure's LIBS through to extension m http://bugs.python.org/issue11370 opened by jszakmeister #11371: Localization of error messages in getopt http://bugs.python.org/issue11371 opened by gruszczy #11373: Fix 2 new typos in the docs http://bugs.python.org/issue11373 opened by Retro #11374: pkgutil.extend_path do not recognize py{c,o} file http://bugs.python.org/issue11374 opened by Alexandre.Badez #11375: urllib2 over SOCKS doesn't use proxy for DNS http://bugs.python.org/issue11375 opened by OJW #11376: Solaris/GCC/shared lib problem http://bugs.python.org/issue11376 opened by dabrahams #11379: Remove "lightweight" from minidom description http://bugs.python.org/issue11379 opened by scoder #11380: "close failed in file object destructor" when "Broken pipe" ha http://bugs.python.org/issue11380 opened by ulidtko #11382: some posix module functions unnecessarily release the GIL http://bugs.python.org/issue11382 opened by neologix #11383: compilation seg faults on insanely large expressions http://bugs.python.org/issue11383 opened by ncoghlan #11385: TextTestRunner methods are not documented http://bugs.python.org/issue11385 opened by techtonik #11387: Tkinter, callback functions http://bugs.python.org/issue11387 opened by Nikolay.Fomichev #11388: Implement MutableSequence.clear() http://bugs.python.org/issue11388 opened by eli.bendersky #11389: unittest: no way to control verbosity of doctests from cmd http://bugs.python.org/issue11389 opened by techtonik #11390: doctest: add cmdline parameters http://bugs.python.org/issue11390 opened by techtonik #11391: mmap write segfaults if PROT_WRITE bit is not set in prot http://bugs.python.org/issue11391 opened by neologix #11392: Turtle - better explain 'chaos' demo http://bugs.python.org/issue11392 opened by sandro.tosi #11393: Integrate faulthandler module into Python 3.3 http://bugs.python.org/issue11393 opened by haypo #11394: No Tools/demo on Windows http://bugs.python.org/issue11394 opened by techtonik #11395: print(s) fails on Windows with long strings http://bugs.python.org/issue11395 opened by casevh #11340: test_distutils fails http://bugs.python.org/issue11340 opened by anikom15 Most recent 15 issues with no replies (15) ========================================== #11392: Turtle - better explain 'chaos' demo http://bugs.python.org/issue11392 #11390: doctest: add cmdline parameters http://bugs.python.org/issue11390 #11389: unittest: no way to control verbosity of doctests from cmd http://bugs.python.org/issue11389 #11387: Tkinter, callback functions http://bugs.python.org/issue11387 #11385: TextTestRunner methods are not documented http://bugs.python.org/issue11385 #11380: "close failed in file object destructor" when "Broken pipe" ha http://bugs.python.org/issue11380 #11376: Solaris/GCC/shared lib problem http://bugs.python.org/issue11376 #11375: urllib2 over SOCKS doesn't use proxy for DNS http://bugs.python.org/issue11375 #11367: xml.etree.ElementTree.find(all): docs are wrong http://bugs.python.org/issue11367 #11361: suggestion for os.kill(pid,CTRL_C_EVENT) http://bugs.python.org/issue11361 #11343: Make errors due to full parser stack identifiable http://bugs.python.org/issue11343 #11337: Nothing refers to footnote [1] on page "6. Simple Statements" http://bugs.python.org/issue11337 #11319: Command line option -t (and -tt) does not work for a particula http://bugs.python.org/issue11319 #11312: Confusing sentence in file.readline() doc http://bugs.python.org/issue11312 #11310: Document byte[s|array]() and byte[s|array](count) in docstring http://bugs.python.org/issue11310 Most recent 15 issues waiting for review (15) ============================================= #11392: Turtle - better explain 'chaos' demo http://bugs.python.org/issue11392 #11391: mmap write segfaults if PROT_WRITE bit is not set in prot http://bugs.python.org/issue11391 #11382: some posix module functions unnecessarily release the GIL http://bugs.python.org/issue11382 #11374: pkgutil.extend_path do not recognize py{c,o} file http://bugs.python.org/issue11374 #11373: Fix 2 new typos in the docs http://bugs.python.org/issue11373 #11371: Localization of error messages in getopt http://bugs.python.org/issue11371 #11370: Fix distutils to carry configure's LIBS through to extension m http://bugs.python.org/issue11370 #11365: Integrate Buildroot patches (cross-compilation) http://bugs.python.org/issue11365 #11364: Move from distutils.sysconfig to sysconfig in test_osx_env http://bugs.python.org/issue11364 #11363: Curses - add missing functions to doc http://bugs.python.org/issue11363 #11361: suggestion for os.kill(pid,CTRL_C_EVENT) http://bugs.python.org/issue11361 #11357: Add support for PEP 381 -- Mirror Authenticity http://bugs.python.org/issue11357 #11355: os.mkdir() and os.mkdirat() don't apply SUID/SGID permissions http://bugs.python.org/issue11355 #11354: argparse: nargs could accept range of options count http://bugs.python.org/issue11354 #11350: __setitem__()'s problem of dbm.dumb object pointed out by comm http://bugs.python.org/issue11350 Top 10 most discussed issues (10) ================================= #9276: pickle should support methods http://bugs.python.org/issue9276 15 msgs #11342: ResourceWarning: unclosed file <_io.TextIOWrapper http://bugs.python.org/issue11342 15 msgs #11395: print(s) fails on Windows with long strings http://bugs.python.org/issue11395 13 msgs #8841: getopt errors should be specialized http://bugs.python.org/issue8841 10 msgs #11382: some posix module functions unnecessarily release the GIL http://bugs.python.org/issue11382 10 msgs #11284: slow close file descriptors in subprocess, popen2, os.popen* http://bugs.python.org/issue11284 8 msgs #11321: 9th import of module _pickle always crashes http://bugs.python.org/issue11321 8 msgs #11299: Allow deepcopying and pickling paused generators http://bugs.python.org/issue11299 7 msgs #11332: Increase logging/__init__.py coverage to 97% http://bugs.python.org/issue11332 7 msgs #11335: Memory leak after key function failure in sort http://bugs.python.org/issue11335 7 msgs Issues closed (62) ================== #4761: create Python wrappers for openat() and others http://bugs.python.org/issue4761 closed by pitrou #6064: Add "daemon" argument to threading.Thread constructor http://bugs.python.org/issue6064 closed by pitrou #7322: Socket timeout can cause file-like readline() method to lose d http://bugs.python.org/issue7322 closed by pitrou #8594: Add a "source_address" option to ftplib http://bugs.python.org/issue8594 closed by giampaolo.rodola #8923: Remove unused "errors" argument from _PyUnicode_AsDefaultEncod http://bugs.python.org/issue8923 closed by haypo #8967: Create PyErr_GetWindowsMessage() function http://bugs.python.org/issue8967 closed by haypo #9197: Crash when importing an extension after Py_Initialize, Py_Fina http://bugs.python.org/issue9197 closed by ncoghlan #9373: pulldom has low code coverage http://bugs.python.org/issue9373 closed by benjamin.peterson #9795: nntplib.NNTP should support the context protocol http://bugs.python.org/issue9795 closed by giampaolo.rodola #9931: test_ttk_guionly hangs on XP5 http://bugs.python.org/issue9931 closed by pitrou #10228: Refleak run of test_dbm fails when several dbm modules are ava http://bugs.python.org/issue10228 closed by orsenthil #10338: test_lib2to3 failure on buildbot x86 debian parallel 3.x: node http://bugs.python.org/issue10338 closed by haypo #10755: Add posix.fdlistdir http://bugs.python.org/issue10755 closed by pitrou #10784: os.getpriority() and os.setpriority() http://bugs.python.org/issue10784 closed by giampaolo.rodola #10829: PyUnicode_FromFormatV() bugs with "%" and "%%" format strings http://bugs.python.org/issue10829 closed by haypo #10831: PyUnicode_FromFormatV() doesn't support %li, %lli, %zi http://bugs.python.org/issue10831 closed by haypo #10866: Add sethostname() http://bugs.python.org/issue10866 closed by pitrou #10900: bz2 module fails to uncompress large files http://bugs.python.org/issue10900 closed by pitrou #10911: cgi: add more tests http://bugs.python.org/issue10911 closed by haypo #10956: file.write and file.read don't handle EINTR http://bugs.python.org/issue10956 closed by pitrou #11114: TextIOWrapper.tell extremely slow http://bugs.python.org/issue11114 closed by pitrou #11140: threading.Lock().release() raises _thread.error, not RuntimeEr http://bugs.python.org/issue11140 closed by pitrou #11172: Avoid '.' as runpath on AIX http://bugs.python.org/issue11172 closed by pitrou #11177: Set asyncore create_socket default argument http://bugs.python.org/issue11177 closed by giampaolo.rodola #11214: test_asyncore fails on AIX http://bugs.python.org/issue11214 closed by giampaolo.rodola #11246: PyUnicode_FromFormat("%V") decodes the byte string from ISO-88 http://bugs.python.org/issue11246 closed by haypo #11250: 2to3 truncates files at formfeed character http://bugs.python.org/issue11250 closed by benjamin.peterson #11265: asyncore does not check for EAGAIN and EPIPE errno http://bugs.python.org/issue11265 closed by giampaolo.rodola #11291: poplib suppresses exception on QUIT http://bugs.python.org/issue11291 closed by giampaolo.rodola #11293: Distutils - read the file when using it in long_description http://bugs.python.org/issue11293 closed by eric.araujo #11297: Make ChainMap() public in the collections module. http://bugs.python.org/issue11297 closed by rhettinger #11303: b'x'.decode('latin1') is much slower than b'x'.decode('latin-1 http://bugs.python.org/issue11303 closed by belopolsky #11306: mailbox should test for errno.EROFS http://bugs.python.org/issue11306 closed by r.david.murray #11307: re engine exhaustively explores more than necessary http://bugs.python.org/issue11307 closed by terry.reedy #11314: Subprocess suffers 40% process creation overhead penalty http://bugs.python.org/issue11314 closed by haypo #11318: Python 3.2 FAQ example code typo? http://bugs.python.org/issue11318 closed by georg.brandl #11323: os.sendfile() bo failure on solaris http://bugs.python.org/issue11323 closed by pitrou #11325: compilation warning with sendfile http://bugs.python.org/issue11325 closed by pitrou #11326: connect_ex() implementation missing for SSL sockets http://bugs.python.org/issue11326 closed by pitrou #11327: Running test_time.py in python27 caused python to unexpectedly http://bugs.python.org/issue11327 closed by brett.cannon #11328: NESTED WHILE CYCLES ERROR http://bugs.python.org/issue11328 closed by pitrou #11330: logging doesn't support format flags for asctime keyword, but http://bugs.python.org/issue11330 closed by vinay.sajip #11331: unused line in the logging-cookbook example http://bugs.python.org/issue11331 closed by vinay.sajip #11334: Failed to build _multiprocessing on Mac OS X http://bugs.python.org/issue11334 closed by dev66 #11336: Typo in patch.rst http://bugs.python.org/issue11336 closed by brett.cannon #11341: test_os fails http://bugs.python.org/issue11341 closed by pitrou #11345: Fix a link in library/json docs http://bugs.python.org/issue11345 closed by benjamin.peterson #11348: test_os: test_set_get_priority() fails when high niceness is s http://bugs.python.org/issue11348 closed by giampaolo.rodola #11351: Mac OS X os.sendfile() http://bugs.python.org/issue11351 closed by sdaoden #11353: Python 2.7.1 cannot be built by 2.x baselines of Python http://bugs.python.org/issue11353 closed by pitrou #11356: Include module name on ImportError http://bugs.python.org/issue11356 closed by georg.brandl #11358: Please replace the use of pickle in multiprocessing with json. http://bugs.python.org/issue11358 closed by r.david.murray #11359: Please replace the use of pickle in subprocess with json. http://bugs.python.org/issue11359 closed by ezio.melotti #11360: In documentation of getopt, advertise argparse instead of optp http://bugs.python.org/issue11360 closed by eric.araujo #11372: Remove xrange from argparse docs http://bugs.python.org/issue11372 closed by fdrake #11377: Deprecate (remove?) platform.popen() http://bugs.python.org/issue11377 closed by haypo #11378: os.path.join when second argument starts with '/' (linux/unix http://bugs.python.org/issue11378 closed by amaury.forgeotdarc #11381: pretending the "not" operator is a function behaves surprising http://bugs.python.org/issue11381 closed by georg.brandl #11384: Deprecate, remove or document (correctly) os.popen http://bugs.python.org/issue11384 closed by haypo #11386: Fix exception thrown by bytearray.pop() for empty bytearrays http://bugs.python.org/issue11386 closed by eli.bendersky #11396: add format handler to bytes object http://bugs.python.org/issue11396 closed by eric.smith #1720250: PyGILState_Ensure does not acquires GIL http://bugs.python.org/issue1720250 closed by pitrou

1 0

16:30-tol lesz a pajton
by Kálmán Gergely 04 Mar '11

04 Mar '11

DD nagytargyalo ebben az idoben. Bocsi a keveresert. synapse

1 0

Integrate the faulthandler module into Python 3.3?
by Victor Stinner 04 Mar '11

04 Mar '11

Hi, Some months ago, I proposed a patch to display the Python backtrace on a segfault. The API was not stable, it was too for Python 3.2, and it was enabled by default. I wrote a module instead of a patch so it can be used on any version of Python, and it has a better API. And now I would like to include the module into Python directly to use it more easily. See http://bugs.python.org/issue11393 for the details. The module: https://github.com/haypo/faulthandler It is now possible to dump the backtrace on an user signal (eg. SIGUSR1) or after a delay in seconds (it may be useful for buildbot hangs without user interaction). You can choose to display the current thread or all threads, and in which file the trace is written. So, what do you think? Victor

6 11

Re: [Python-Dev] [Python-checkins] r88729 - python/branches/py3k/Modules/posixmodule.c
by Nick Coghlan 04 Mar '11

04 Mar '11

On Fri, Mar 4, 2011 at 2:10 AM, giampaolo.rodola <python-checkins(a)python.org> wrote: > Author: giampaolo.rodola > Date: Thu Mar 3 17:10:51 2011 > New Revision: 88729 > > Log: > Issue 11351 - apply patch by Steffen Daode Nurpmeso which should fix TestSendfile.test_headers failure on OSX. > > Modified: > python/branches/py3k/Modules/posixmodule.c NEWS entry and a new name in ACKS? (the query regarding the lack of a NEWS entry applies to your other recent commits as well). Cheers, Nick. -- Nick Coghlan | ncoghlan(a)gmail.com | Brisbane, Australia

2 2