Mailman 3 September 2012 - Python-Dev

"Decimal(2) != float(2)"???
by Jan Kaliszewski Sept. 30, 2012

Sept. 30, 2012

Hello, In http://docs.python.org/release/3.2.3/reference/expressions.html#in we read: "[...] This can create the illusion of non-transitivity between supported cross-type comparisons and unsupported comparisons. For example, Decimal(2) == 2 and 2 == float(2) but Decimal(2) != float(2)." (The same is in the 3.3 docs). But: Python 3.2.3 (default, Sep 10 2012, 18:14:40) [GCC 4.6.3] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import decimal >>> decimal.Decimal(2) == float(2) True Is it a bug in the docs or in Python itself? (I checked that in 3.2, but it may be true for 3.3 as well) Regards. *j

4 3

Re: [Python-Dev] make decimal the default non-integer instead of float?
by Gregory P. Smith Sept. 29, 2012

Sept. 29, 2012

-cc: python-dev +cc: python-ideas On Sat, Sep 29, 2012 at 11:39 AM, Chris Angelico <rosuav(a)gmail.com> wrote: > On Sun, Sep 30, 2012 at 4:26 AM, Brett Cannon <brett(a)python.org> wrote: > > Does this mean we want to re-open the discussion about decimal constants? > > Last time this came up I think we decided that we wanted to wait for > > cdecimal (which is obviously here) and work out how to handle contexts, > the > > syntax, etc. > > Just to throw a crazy idea out: How bad a change would it be to make > decimal actually the default? > > (Caveat: I've not worked with decimal/cdecimal to any real extent and > don't know its limitations etc.) > Painful for existing code, unittests and extension modules. Definitely python-ideas territory (thread moved there with an appropriate subject). I'm not surprised at all that a decimal type can be "fast" in an interpreted language due to the already dominant interpreter overhead. I wish all spreadsheets had used decimals from day one rather than binary floating point (blame Lotus?). Think of the trouble that would have saved the world. -gps

1 0

str.format bug again
by Ben Wolfson Sept. 29, 2012

Sept. 29, 2012

There's a patch for this bug: http://bugs.python.org/issue12014 that needs reviewed. Petri Lehtinen made some (very minor) comments back in May; otherwise it's been neglected. I've brought this up occasionally here in the past; it would be great if someone could just give it a thumbs up or down (or say what needs to be changed, or whatever). -- Ben Wolfson "Human kind has used its intelligence to vary the flavour of drinks, which may be sweet, aromatic, fermented or spirit-based. ... Family and social life also offer numerous other occasions to consume drinks for pleasure." [Larousse, "Drink" entry]

3 3

Deprecating "U" open mode
by Serhiy Storchaka Sept. 28, 2012

Sept. 28, 2012

Since Python 2.3 many open-like functions supports "Universal line mode" (PEP 278 [1]). Since 3.0 (and 2.6) PEP 3116 [2] suggests better alternative -- io.TextWrapper. In issue15204 [3] proposed to deprecate "U" mode in open-like functions. In fact I found only three places where "U" mode is mentioned in the standard library: * builtin open (io.open) (almost ignored); * fileinput (transparently passed to open); * zipfile (inefficient and inconsistent implementation). Is someone uses "U" mode and has objections? [1] http://www.python.org/dev/peps/pep-0278/ [2] http://www.python.org/dev/peps/pep-3116/ [3] http://bugs.python.org/issue15204

2 1

Summary of Python tracker Issues
by Python tracker Sept. 28, 2012

Sept. 28, 2012

ACTIVITY SUMMARY (2012-09-21 - 2012-09-28) 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 3754 (+23) closed 24106 (+61) total 27860 (+84) Open issues with patches: 1662 Issues opened (52) ================== #16000: test_curses should use unittest http://bugs.python.org/issue16000 opened by chris.jerdonek #16001: small ints: cache string representation http://bugs.python.org/issue16001 opened by haypo #16003: thread_PyThread_start_new_thread fails with a generic error http://bugs.python.org/issue16003 opened by kristjan.jonsson #16004: Add `make touch` to 2.7 Makefile http://bugs.python.org/issue16004 opened by chris.jerdonek #16005: smtplib.SMTP().sendmail() and rset() http://bugs.python.org/issue16005 opened by DDarko #16007: Improved Error message for failing re expressions http://bugs.python.org/issue16007 opened by lingster #16009: Json error messages could provide more information about the e http://bugs.python.org/issue16009 opened by Luka.Rahne #16011: "in" should be consistent with return value of __contains__ http://bugs.python.org/issue16011 opened by nparikh #16013: small csv reader bug http://bugs.python.org/issue16013 opened by arigo #16023: IDLE freezes on ^5 or ^6 (Un-)Tabify Region with OS X Cocoa Tk http://bugs.python.org/issue16023 opened by szellmeyer #16024: Doc cleanup regarding path=fd, dir_fd, follow_symlinks, etc http://bugs.python.org/issue16024 opened by larry #16025: Minor corrections to the zipfile documentation http://bugs.python.org/issue16025 opened by storchaka #16026: csv.DictReader argument names documented incorrectly http://bugs.python.org/issue16026 opened by petere #16027: pkgutil doesn't support frozen modules http://bugs.python.org/issue16027 opened by lemburg #16029: pickle.dumps(xrange(sys.maxsize)) produces xrange(0) http://bugs.python.org/issue16029 opened by akira #16030: xrange repr broken for large arguments http://bugs.python.org/issue16030 opened by mark.dickinson #16034: bz2 module appears slower in Python 3.x versus Python 2.x http://bugs.python.org/issue16034 opened by victorhooi #16036: simplify int() signature docs http://bugs.python.org/issue16036 opened by chris.jerdonek #16037: httplib: header parsing is not delimited http://bugs.python.org/issue16037 opened by christian.heimes #16038: ftplib: unlimited readline() from connection http://bugs.python.org/issue16038 opened by christian.heimes #16039: imaplib: unlimited readline() from connection http://bugs.python.org/issue16039 opened by christian.heimes #16040: nntplib: unlimited readline() from connection http://bugs.python.org/issue16040 opened by christian.heimes #16041: poplib: unlimited readline() from connection http://bugs.python.org/issue16041 opened by christian.heimes #16042: smtplib: unlimited readline() from connection http://bugs.python.org/issue16042 opened by christian.heimes #16043: xmlrpc: gzip_decode has unlimited read() http://bugs.python.org/issue16043 opened by christian.heimes #16044: xml.etree.ElementTree.Element: append method iterator param is http://bugs.python.org/issue16044 opened by kirpit #16045: add more unit tests for built-in int() http://bugs.python.org/issue16045 opened by chris.jerdonek #16047: Tools/freeze no longer works in Python 3 http://bugs.python.org/issue16047 opened by lemburg #16048: Tutorial-classes-remarks: replace paragragh http://bugs.python.org/issue16048 opened by terry.reedy #16049: Create abstract base classes by inheritance rather than a dire http://bugs.python.org/issue16049 opened by rhettinger #16050: ctypes: callback from C++ to Python fails with Illegal Instruc http://bugs.python.org/issue16050 opened by Opilki_Inside #16053: "strict" parameter is not documented in csv module http://bugs.python.org/issue16053 opened by storchaka #16055: incorrect error text for int(base=1000, x='1') http://bugs.python.org/issue16055 opened by chris.jerdonek #16056: shadowed test names in std lib regression tests http://bugs.python.org/issue16056 opened by xdegaye #16057: Subclasses of JSONEncoder should not be insturcted to call JSO http://bugs.python.org/issue16057 opened by Justin.Lebar #16058: ConfigParser no longer deepcopy compatible in 2.7 http://bugs.python.org/issue16058 opened by rbcollins #16059: Serialize MD5 computation-state and resume later http://bugs.python.org/issue16059 opened by Ye.Yuan #16061: performance regression in string replace for 3.3 http://bugs.python.org/issue16061 opened by BreamoreBoy #16062: Socket closed prematurely in httplib for https http://bugs.python.org/issue16062 opened by ABR #16063: HMAC trans_5C is a string, causing a TypeError http://bugs.python.org/issue16063 opened by Adam.Glenn #16065: Python/distutils setup.py: passing --prefix / makes --root ign http://bugs.python.org/issue16065 opened by arschficker #16066: Truncated POST data in CGI script on Windows 7 http://bugs.python.org/issue16066 opened by Alexander.Martin #16067: UAC prompt for installation shows temporary file name http://bugs.python.org/issue16067 opened by breyed #16070: Structure and native Structure (LittleEndianStructure on Windo http://bugs.python.org/issue16070 opened by hct #16071: fix link to email.message.Message in mailbox docs http://bugs.python.org/issue16071 opened by chris.jerdonek #16074: bad error in rename http://bugs.python.org/issue16074 opened by v+python #16076: xml.etree.ElementTree.Element is no longer pickleable http://bugs.python.org/issue16076 opened by einarfd #16077: fix code example in docs for built-in reduce() http://bugs.python.org/issue16077 opened by chris.jerdonek #16079: list duplicate test names with patchcheck http://bugs.python.org/issue16079 opened by xdegaye #16081: Fix compile warnings in thread_pthread.h http://bugs.python.org/issue16081 opened by bbrazil #16082: xml.etree.ElementTree.tostringlist does conform to it's docume http://bugs.python.org/issue16082 opened by einarfd #16083: HTTPServer does not correctly handle bad headers http://bugs.python.org/issue16083 opened by maker Most recent 15 issues with no replies (15) ========================================== #16081: Fix compile warnings in thread_pthread.h http://bugs.python.org/issue16081 #16079: list duplicate test names with patchcheck http://bugs.python.org/issue16079 #16076: xml.etree.ElementTree.Element is no longer pickleable http://bugs.python.org/issue16076 #16074: bad error in rename http://bugs.python.org/issue16074 #16071: fix link to email.message.Message in mailbox docs http://bugs.python.org/issue16071 #16070: Structure and native Structure (LittleEndianStructure on Windo http://bugs.python.org/issue16070 #16067: UAC prompt for installation shows temporary file name http://bugs.python.org/issue16067 #16063: HMAC trans_5C is a string, causing a TypeError http://bugs.python.org/issue16063 #16058: ConfigParser no longer deepcopy compatible in 2.7 http://bugs.python.org/issue16058 #16057: Subclasses of JSONEncoder should not be insturcted to call JSO http://bugs.python.org/issue16057 #16053: "strict" parameter is not documented in csv module http://bugs.python.org/issue16053 #16047: Tools/freeze no longer works in Python 3 http://bugs.python.org/issue16047 #16041: poplib: unlimited readline() from connection http://bugs.python.org/issue16041 #16040: nntplib: unlimited readline() from connection http://bugs.python.org/issue16040 #16039: imaplib: unlimited readline() from connection http://bugs.python.org/issue16039 Most recent 15 issues waiting for review (15) ============================================= #16083: HTTPServer does not correctly handle bad headers http://bugs.python.org/issue16083 #16081: Fix compile warnings in thread_pthread.h http://bugs.python.org/issue16081 #16079: list duplicate test names with patchcheck http://bugs.python.org/issue16079 #16056: shadowed test names in std lib regression tests http://bugs.python.org/issue16056 #16055: incorrect error text for int(base=1000, x='1') http://bugs.python.org/issue16055 #16050: ctypes: callback from C++ to Python fails with Illegal Instruc http://bugs.python.org/issue16050 #16042: smtplib: unlimited readline() from connection http://bugs.python.org/issue16042 #16036: simplify int() signature docs http://bugs.python.org/issue16036 #16034: bz2 module appears slower in Python 3.x versus Python 2.x http://bugs.python.org/issue16034 #16029: pickle.dumps(xrange(sys.maxsize)) produces xrange(0) http://bugs.python.org/issue16029 #16025: Minor corrections to the zipfile documentation http://bugs.python.org/issue16025 #16024: Doc cleanup regarding path=fd, dir_fd, follow_symlinks, etc http://bugs.python.org/issue16024 #16013: small csv reader bug http://bugs.python.org/issue16013 #16004: Add `make touch` to 2.7 Makefile http://bugs.python.org/issue16004 #16001: small ints: cache string representation http://bugs.python.org/issue16001 Top 10 most discussed issues (10) ================================= #8425: a -= b should be fast if a is a small set and b is a large set http://bugs.python.org/issue8425 14 msgs #10044: small int optimization http://bugs.python.org/issue10044 10 msgs #15958: bytes.join() should allow arbitrary buffer objects http://bugs.python.org/issue15958 9 msgs #15964: SyntaxError in asdl when building 2.7 with system Python 3 http://bugs.python.org/issue15964 9 msgs #16029: pickle.dumps(xrange(sys.maxsize)) produces xrange(0) http://bugs.python.org/issue16029 9 msgs #16036: simplify int() signature docs http://bugs.python.org/issue16036 9 msgs #16055: incorrect error text for int(base=1000, x='1') http://bugs.python.org/issue16055 8 msgs #16056: shadowed test names in std lib regression tests http://bugs.python.org/issue16056 8 msgs #16001: small ints: cache string representation http://bugs.python.org/issue16001 7 msgs #16005: smtplib.SMTP().sendmail() and rset() http://bugs.python.org/issue16005 7 msgs Issues closed (58) ================== #2126: BaseHTTPServer.py fails long POST from IE http://bugs.python.org/issue2126 closed by pitrou #2600: BindingHTTPConnectionWithTimeout and BindingHTTPHandlerWithTim http://bugs.python.org/issue2600 closed by pitrou #6027: test_xmlrpc_net fails when the ISP returns "302 Found" http://bugs.python.org/issue6027 closed by ezio.melotti #7844: Add -3 warning for absolute imports. http://bugs.python.org/issue7844 closed by mark.dickinson #10980: http.server Header Unicode Bug http://bugs.python.org/issue10980 closed by pitrou #11103: Python 3.2 installer doesn't register file extensions on Windo http://bugs.python.org/issue11103 closed by serwy #12376: unittest.TextTestResult.__init__ does not pass on its init arg http://bugs.python.org/issue12376 closed by michael.foord #13263: Group some os functions in submodules http://bugs.python.org/issue13263 closed by ezio.melotti #14295: PEP 417: adding mock module http://bugs.python.org/issue14295 closed by michael.foord #15270: "Economy of Expression" section outdated http://bugs.python.org/issue15270 closed by rhettinger #15323: Provide target name in output message when Mock.assert_called_ http://bugs.python.org/issue15323 closed by python-dev #15379: Charmap decoding of no-BMP characters http://bugs.python.org/issue15379 closed by pitrou #15593: urlparse.parse_qs documentation wrong re: urlencode http://bugs.python.org/issue15593 closed by python-dev #15815: Add numerator to ZeroDivisionError messages http://bugs.python.org/issue15815 closed by terry.reedy #15882: _decimal.Decimal constructed from tuple http://bugs.python.org/issue15882 closed by georg.brandl #15895: PyRun_SimpleFileExFlags() can leak a FILE pointer http://bugs.python.org/issue15895 closed by georg.brandl #15900: Memory leak in PyUnicode_TranslateCharmap() http://bugs.python.org/issue15900 closed by georg.brandl #15925: email.utils.parsedate() and email.utils.parsedate_tz() should http://bugs.python.org/issue15925 closed by georg.brandl #15926: Segmentation fault after multiple reinitializations http://bugs.python.org/issue15926 closed by georg.brandl #15935: clarify argparse docs re: add_argument() type and default argu http://bugs.python.org/issue15935 closed by barry #15949: docs.python.org not getting updated http://bugs.python.org/issue15949 closed by ezio.melotti #15953: Incorrect some fields declaration in the PyTypeObject document http://bugs.python.org/issue15953 closed by python-dev #15969: faulthandler: drop S from dump_tracebacks_later() function nam http://bugs.python.org/issue15969 closed by georg.brandl #15973: Segmentation fault on timezone comparison http://bugs.python.org/issue15973 closed by georg.brandl #15978: asyncore: included batteries don't fit http://bugs.python.org/issue15978 closed by benjamin.peterson #15993: Windows: 3.3.0-rc2.msi: test_buffer fails http://bugs.python.org/issue15993 closed by loewis #16002: AIX buildbot compile errors http://bugs.python.org/issue16002 closed by skrah #16006: Spelling mistake in PEP 213 http://bugs.python.org/issue16006 closed by ezio.melotti #16008: [OS bug] DragonFly buildbot locale failures http://bugs.python.org/issue16008 closed by skrah #16010: Some Unicode in identifiers improperly rejected http://bugs.python.org/issue16010 closed by r.david.murray #16012: pyexpat: Argument for UseForeignDTD should be optional http://bugs.python.org/issue16012 closed by christian.heimes #16014: Incomplete link in docs http://bugs.python.org/issue16014 closed by ezio.melotti #16015: Incorrect startup header in tutorial http://bugs.python.org/issue16015 closed by chris.jerdonek #16016: Maybe make one author in ipaddress HOWTO http://bugs.python.org/issue16016 closed by ezio.melotti #16017: "urls" should be capital in tutorial http://bugs.python.org/issue16017 closed by ezio.melotti #16018: Impossible ... line in tutorial http://bugs.python.org/issue16018 closed by ezio.melotti #16019: Python highlighting in tutorial;should be text highlighting http://bugs.python.org/issue16019 closed by ezio.melotti #16020: Missing >>> in Python code example http://bugs.python.org/issue16020 closed by ezio.melotti #16021: In a table in PEP 235, brrrrrrrr should be N/A http://bugs.python.org/issue16021 closed by mark.dickinson #16022: Mistake in "What's New in Python 3.3" http://bugs.python.org/issue16022 closed by python-dev #16028: break in finally discards exception http://bugs.python.org/issue16028 closed by mark.dickinson #16031: relative import headaches http://bugs.python.org/issue16031 closed by ncoghlan #16032: IDLE 3.2 is crashing multiple times http://bugs.python.org/issue16032 closed by ned.deily #16033: IDLE crashes when debugging and saving http://bugs.python.org/issue16033 closed by serwy #16035: Segmentation fault in test suite of apsw http://bugs.python.org/issue16035 closed by georg.brandl #16046: python -O does not find *.pyo files http://bugs.python.org/issue16046 closed by georg.brandl #16051: Documentation missing pipes.quote() http://bugs.python.org/issue16051 closed by r.david.murray #16052: Typo in "What's new in 3.3" http://bugs.python.org/issue16052 closed by python-dev #16054: IDLE Preferences Crashes on click http://bugs.python.org/issue16054 closed by ned.deily #16060: Double decref and dereferencing after decref in int() http://bugs.python.org/issue16060 closed by georg.brandl #16064: unittest -m claims executable is "python", not "python3" http://bugs.python.org/issue16064 closed by python-dev #16068: with statement executes type(obj).__exit__ rather than getattr http://bugs.python.org/issue16068 closed by Paul.Wiseman #16069: packaging shows up on docs.python.org/dev http://bugs.python.org/issue16069 closed by georg.brandl #16072: fix documentation of string.replace() signature http://bugs.python.org/issue16072 closed by asvetlov #16073: fix map() statement in list comprehension example http://bugs.python.org/issue16073 closed by chris.jerdonek #16075: incoming.cia hook error when pushing commits http://bugs.python.org/issue16075 closed by georg.brandl #16078: Calendar.leapdays(y1,y2) bug http://bugs.python.org/issue16078 closed by larry #16080: test_decimal causes other tests to fail with LC_ALL=fr_FR http://bugs.python.org/issue16080 closed by skrah

1 0

Re: [Python-Dev] [Distutils] distlib updated with resources API
by Tarek Ziadé Sept. 27, 2012

Sept. 27, 2012

On 9/28/12 12:55 AM, Antoine Pitrou wrote: >> Last but not least, distlib is the plan forward endorsed by python-dev, > Is it? I haven't seen a PEP or an official decision about that. Just because > someone proposed it on a mailing-list doesn't mean it is "endorsed by > python-dev". We discussed about this with Vinay, Nick and al on python-dev, based on Nick's document that describes what 'distlib' is. The document has changed since then, http://python-notes.boredomandlaziness.org/en/latest/pep_ideas/core_packagi… But the idea was to create a subset of 4 or 5 modules that implement the various PEPs. Vinay started to work on this and made progress. When I said "endorsed", I mean that most of the people in python-dev that care about packaging agreed or did not disagree. Now, if you disagree please say it. Or if you need an official decision, we need to first declare who is the current packaging BDFL maybe ? And since you seem interested in the topic maybe you could take that role ? Cheers Tarek

2 1

Re: [Python-Dev] [Python-checkins] cpython: Add a few entries to whatsnew/3.3.rst.
by Walter Dörwald Sept. 27, 2012

Sept. 27, 2012

On 26.09.12 16:43, ezio.melotti wrote: > http://hg.python.org/cpython/rev/36f61661f71e > changeset: 79194:36f61661f71e > user: Ezio Melotti <ezio.melotti(a)gmail.com> > date: Wed Sep 26 17:43:23 2012 +0300 > summary: > Add a few entries to whatsnew/3.3.rst. > [...] > + > +A new :data:`~html.entities.html5` dictionary that maps HTML5 named character > +references to the equivalent Unicode character(s) (e.g. ``html5['gt;'] == '>'``) > +has been added to the :mod:`html.entities` module. The dictionary is now also > +used by :class:`~html.parser.HTMLParser`. Is there a reason why the trailing ';' is included in the entity names? Servus, Walter

3 2

Re: [Python-Dev] [Python-checkins] cpython: don't depend on __debug__ because it's baked in at freeze time (issue #16046)
by Nick Coghlan Sept. 26, 2012

Sept. 26, 2012

On Wed, Sep 26, 2012 at 1:23 AM, benjamin.peterson <python-checkins(a)python.org> wrote: > If debug_override is not None, then it must be a boolean and is taken as > - the value of __debug__ instead. > + the value of bool(sys.flags.optimize) instead. s/bool(sys.flags.optimize)/"not sys.flags.optimize"/? It may be better to just rephrase this sentence entirely to better account for the fact that we're now checking the runtime sys.flags.optimize value rather than the compile time __debug__ value. Cheers, Nick. -- Nick Coghlan | ncoghlan(a)gmail.com | Brisbane, Australia

2 1

Better HTTP 1.1 support in http.server?
by Antoine Pitrou Sept. 25, 2012

Sept. 25, 2012

Hello, I was wondering if it would be worth providing better HTTP 1.1 support in http.server. The way I envision it, there would be a separate HTTP11RequestHandler which would provide: - a smart wfile with automatic chunk encoding (which relieves the API user from manually handling chunk encoding or content length calculation) - keep-alive enabled by default - HTTP 1.1 by default (BaseHTTPRequestHandler defaults to 1.0) What do you think? Regards Antoine. -- Software development and contracting: http://pro.pitrou.net

6 5

3.3 release
by Yury Selivanov Sept. 22, 2012

Sept. 22, 2012

Hello, What's the status of 3.3? Is it going to be released tomorrow? (I see that we still have a few release-blockers in the tracker, and the whatsnew section is probably due to be updated by Raymond...) - Yury

2 1